---
schema_version: 7
rubric_version: "0.3"
canonical_url: "https://podium.do/compare/agents/products/martinloop/"
data_url: "https://podium.do/compare/agents/products/martinloop/data.json"
generated_at: 2026-11-22
oldest_checked_at: 2026-11-16
last_checked_at: 2026-11-16
publisher: "Podium"
---

# MartinLoop: ADE findings

> Podium publishes this catalog and sells an agent development environment. Podium chooses the products, criteria, sources and wording, so treat these pages as vendor-authored research. Check important claims against the linked primary sources. Sourced findings follow the same evidence rules for Podium and its competitors. Facts labeled Podium team research come from Podium's own team and have no linked source. None of this removes the conflict of interest.

> Based on data from adecomparison.com by Podium (https://adecomparison.com; source repository https://github.com/madeinorbit/ade-comparison, snapshot 4322f0c of 2026-11-22), used under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Source links and check dates are kept. Podium arranges the findings into pairs, applies its publication checks, and added its own research where marked "Podium team research". Data license terms: https://github.com/madeinorbit/ade-comparison/blob/4322f0c035f303791413639d1462a138443d8c1e/DATA-LICENSE.md

50 of 51 criteria have a sourced finding; 37 were also cross-checked by a second agent. Source checks: 2026-11-16. A missing sourced finding is not evidence that a capability is absent.

- Website: https://martinloop.com
- Canonical HTML: https://podium.do/compare/agents/products/martinloop/
- Machine-readable JSON: https://podium.do/compare/agents/products/martinloop/data.json
- Rubric: https://podium.do/compare/agents/rubric/index.md
- Comparisons with MartinLoop on podium.do: 1 (listed in the JSON and HTML)
- Compare MartinLoop with any other product on adecomparison.com by Podium: https://adecomparison.com/compare/?products=martinloop

## Access and agent choice

### ACC-01: Client access

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Terminal CLI (`martin-loop`/`martin`), a local stdio/HTTP MCP server and a TypeScript SDK; release validated on Windows, macOS and Linux. No desktop, mobile or editor app; a closed hosted dashboard exists on the Pro plan.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+), @martinloop/mcp server (stdio and HTTP transports), Hosted dashboard / Atlas (closed, Pro plan); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- desktop_os: No desktop app; the Node CLI is validated by hosted Windows, macOS and Linux release matrices.
- mobile_apps: No mobile app in the repository or product pages; bounded search found no iOS/Android code.
- browser: The closed hosted Pro dashboard (martinloop.com) shows runs, receipts and cost history; the open-source core has no browser UI and the hosted client cannot be inspected.
- mobile_browser: Responsive behavior of the closed hosted dashboard is not documented; the open-source core has no browser UI.
- editor_extensions: No own editor extension; the MCP server installs into VS Code/Cursor/Codex/Claude/Gemini hosts and a Claude plugin manifest ships.
- task_actions: Start (`run`), review (`dossier`, `share`), resume (`resume`/`continue`) and, over local stdio MCP, pause/continue/cancel; approvals are flags set before a run and there is no live steering input.
- notifications: No notification code in the pinned packages; terminal output and an optional Arcade game only.
- cross_device_continuity: Runs persist in the local runs root (~/.martin/runs); optional hosted sync can upload receipts for viewing on another device but does not move execution.

- desktop_os — Desktop os [partial]: No desktop app; the Node CLI is validated by hosted Windows, macOS and Linux release matrices. Sources: martinloop:release-ci, martinloop:readme.
- mobile_apps — Mobile apps [unsupported]: No mobile app in the repository or product pages; bounded search found no iOS/Android code. Sources: martinloop:srch-clients, martinloop:readme.
- browser — Browser [partial]: The closed hosted Pro dashboard (martinloop.com) shows runs, receipts and cost history; the open-source core has no browser UI and the hosted client cannot be inspected. Sources: martinloop:pricing, martinloop:llms.
- mobile_browser — Mobile browser [unknown]: Responsive behavior of the closed hosted dashboard is not documented; the open-source core has no browser UI. Sources: martinloop:pricing.
- cli — Cli [supported]: `martin-loop`/`martin` CLI with start, doctor, run, preflight, dossier, resume, share, runs, swarm, sync and mcp commands. Sources: martinloop:cli-ref, martinloop:readme.
- editor_extensions — Editor extensions [unsupported]: No own editor extension; the MCP server installs into VS Code/Cursor/Codex/Claude/Gemini hosts and a Claude plugin manifest ships. Sources: martinloop:readme, martinloop:plugin, martinloop:srch-clients.
- task_actions — Task actions [partial]: Start (`run`), review (`dossier`, `share`), resume (`resume`/`continue`) and, over local stdio MCP, pause/continue/cancel; approvals are flags set before a run and there is no live steering input. Sources: martinloop:cli-ref, martinloop:mcp-tools, martinloop:leash.
- notifications — Notifications [unsupported]: No notification code in the pinned packages; terminal output and an optional Arcade game only. Sources: martinloop:srch-notify, martinloop:readme.
- cross_device_continuity — Cross device continuity [partial]: Runs persist in the local runs root (~/.martin/runs); optional hosted sync can upload receipts for viewing on another device but does not move execution. Sources: martinloop:readme, martinloop:cli-ref, martinloop:hosted-handoff.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:readme, martinloop:cli-ref, martinloop:mcp-tools, martinloop:release-ci, martinloop:pricing, martinloop:srch-clients, martinloop:llms, martinloop:plugin, martinloop:leash, martinloop:srch-notify, martinloop:hosted-handoff

### ACC-02: Agent and model choice

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Native adapters for Codex CLI, Claude Code CLI and Gemini CLI, plus an OpenAI-compatible HTTP adapter for any compatible endpoint; the product controls sessions through those adapters and has an ordered provider fallback chain.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+), martin-loop TypeScript SDK and adapters; version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- session_integration: Native adapters launch the agent CLIs and observe workspace changes; the openai adapter embeds a constrained structured-edit loop; no session API control beyond process launch.
- mixed_harness_operation: An explicit ordered `--fallback-engine` chain can move a run to another engine after provider failure; swarm agents mix is not documented for different harnesses in one run.

- harnesses — Harnesses [supported]: Native CLI adapters for Codex, Claude Code and Gemini CLI (`--engine codex|claude|gemini|auto`). Sources: martinloop:models, martinloop:readme.
- model_providers — Model providers [supported]: `--model` passes through unchanged; the `openai` engine reaches OpenAI, OpenRouter, Together, Fireworks, DeepSeek, Kimi, Qwen, Mistral and local Ollama/LM Studio/llama.cpp endpoints. Sources: martinloop:models.
- custom_adapters — Custom adapters [supported]: SDK exports createClaudeCliAdapter, createCodexCliAdapter, createGeminiCliAdapter, createDirectProviderAdapter and createOpenAiCompatibleAdapter; an OpenCode-style adapter example is shipped. Sources: martinloop:readme, martinloop:examples.
- session_integration — Session integration [partial]: Native adapters launch the agent CLIs and observe workspace changes; the openai adapter embeds a constrained structured-edit loop; no session API control beyond process launch. Sources: martinloop:models.
- mixed_harness_operation — Mixed harness operation [partial]: An explicit ordered `--fallback-engine` chain can move a run to another engine after provider failure; swarm agents mix is not documented for different harnesses in one run. Sources: martinloop:changelog, martinloop:models.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:readme, martinloop:models, martinloop:changelog, martinloop:examples

### ACC-03: Subscriptions, API keys and vendor credits

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Uses the host agent CLI's own login (Claude Code, Codex, Gemini) or API keys for the OpenAI-compatible engine; local models work through Ollama/LM Studio; no vendor credits or MartinLoop-held accounts.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- vendor_credits: No MartinLoop-provided model credits; spend is billed by the provider.
- account_switching: Engine is chosen per run with `--engine` and saved with `enable --engine`; no multi-account switcher within one vendor.
- provider_restrictions: Provider terms on automated use of subscriptions are not stated in the repository; they are vendor policy.

- agent_subscriptions — Agent subscriptions [supported]: Native CLI adapters run the authenticated host runtime, so an existing Claude Code, Codex or Gemini login is used; MartinLoop injects no hidden model. Sources: martinloop:readme, martinloop:models.
- api_accounts — Api accounts [supported]: The openai engine takes MARTIN_OPENAI_BASE_URL/API_KEY/MODEL for provider API accounts. Sources: martinloop:models.
- vendor_credits — Vendor credits [unsupported]: No MartinLoop-provided model credits; spend is billed by the provider. Sources: martinloop:readme, martinloop:pricing.
- local_models — Local models [supported]: Local endpoints (Ollama, LM Studio, llama.cpp) via the openai engine; API key optional. Sources: martinloop:models.
- account_switching — Account switching [partial]: Engine is chosen per run with `--engine` and saved with `enable --engine`; no multi-account switcher within one vendor. Sources: martinloop:readme, martinloop:cli-ref.
- vendor_account_requirement — Vendor account requirement [supported]: No MartinLoop account is needed for Core; hosted sync and the dashboard need MARTIN_API_TOKEN. Sources: martinloop:readme, martinloop:cli-ref.
- provider_restrictions — Provider restrictions [unknown]: Provider terms on automated use of subscriptions are not stated in the repository; they are vendor policy. Sources: martinloop:models.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:readme, martinloop:models, martinloop:pricing, martinloop:cli-ref

### ACC-04: Onboarding and project prerequisites

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Documented path: `npx martin-loop start`, `enable --engine`, `demo`, then `run "<task>" --verify "npm test"`; needs Node 20+ and a supported agent CLI. Works on a Git repo; no remote or cloud setup exists.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- no_git: Rollback and change observation rely on Git for the repository root; non-Git directory behavior is not documented (rollback code shells out to git).
- hosted_git: `martin_create_pr` uses the GitHub CLI for PRs; no hosted-Git connection or app is managed by MartinLoop.
- managed_cloud_setup: No managed-cloud execution exists; bounded search found no remote runner.
- byo_vps_onboarding: No VPS installer or remote onboarding flow in the pinned release.

- first_task_setup — First task setup [supported]: start → demo workspace → `run` with --verify, --budget-usd and --max-iterations; `run` auto-checks doctor, session-start and preflight. Sources: martinloop:readme, martinloop:cli-ref.
- no_git — No git [partial]: Rollback and change observation rely on Git for the repository root; non-Git directory behavior is not documented (rollback code shells out to git). Sources: martinloop:rollback, martinloop:secpol.
- local_git_without_remote — Local git without remote [supported]: Runs, rollback boundaries and receipts are local; no remote is required. Sources: martinloop:rollback, martinloop:readme.
- hosted_git — Hosted git [partial]: `martin_create_pr` uses the GitHub CLI for PRs; no hosted-Git connection or app is managed by MartinLoop. Sources: martinloop:prtools.
- prerequisite_guidance — Prerequisite guidance [supported]: Node 20+, an installed and authenticated agent CLI, and a verifier command; `doctor` checks readiness. Sources: martinloop:readme, martinloop:cli-ref.
- managed_cloud_setup — Managed cloud setup [unsupported]: No managed-cloud execution exists; bounded search found no remote runner. Sources: martinloop:readme, martinloop:srch-isolation.
- byo_vps_onboarding — Byo vps onboarding [unsupported]: No VPS installer or remote onboarding flow in the pinned release. Sources: martinloop:readme, martinloop:srch-isolation.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:readme, martinloop:cli-ref, martinloop:secpol, martinloop:rollback, martinloop:prtools, martinloop:srch-isolation

### ACC-05: Programmable access

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

CLI with JSON output, a TypeScript SDK, and an MCP server exposing run, plan, status, pause/continue/cancel, dossier and PR tools over stdio (and an HTTP option); no REST API or webhooks in Core.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+), @martinloop/mcp server (stdio and HTTP transports), martin-loop TypeScript SDK and adapters; version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- api: No hosted REST API in Core; a hosted sync route exists only as a closed native control plane contract.
- webhooks: No inbound or outbound webhook code in the pinned packages.
- authentication: Local CLI/MCP use the OS user (a same-machine trust boundary); hosted sync uses MARTIN_API_TOKEN.
- stability: Pre-1.0 (0.9.0); changelog states no CLI command or MCP tool removed in 0.9.0, no formal API stability policy found.

- cli — Cli [supported]: `martin-loop run/preflight/dossier/runs/swarm/sync` with `--json` machine mode. Sources: martinloop:cli-ref, martinloop:readme.
- api — Api [partial]: No hosted REST API in Core; a hosted sync route exists only as a closed native control plane contract. Sources: martinloop:hosted-handoff.
- sdk — Sdk [supported]: `npm install martin-loop` exports MartinLoop and adapter factories. Sources: martinloop:readme.
- mcp_server — Mcp server [supported]: @martinloop/mcp exposes martin_run, martin_plan, martin_status, martin_pause/cancel/continue, inspection and PR tools. Sources: martinloop:mcp-tools, martinloop:mcp-server.
- webhooks — Webhooks [unsupported]: No inbound or outbound webhook code in the pinned packages. Sources: martinloop:srch-sched.
- operations — Operations [supported]: Create (run/swarm run), inspect (runs/dossier), stop (cancel), pause/continue (MCP) and review (martin_review_pr). Sources: martinloop:mcp-tools, martinloop:cli-ref.
- authentication — Authentication [partial]: Local CLI/MCP use the OS user (a same-machine trust boundary); hosted sync uses MARTIN_API_TOKEN. Sources: martinloop:mcp-tools, martinloop:cli-ref.
- stability — Stability [partial]: Pre-1.0 (0.9.0); changelog states no CLI command or MCP tool removed in 0.9.0, no formal API stability policy found. Sources: martinloop:changelog, martinloop:rel090.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:readme, martinloop:cli-ref, martinloop:mcp-tools, martinloop:mcp-server, martinloop:hosted-handoff, martinloop:srch-sched, martinloop:changelog, martinloop:rel090

### ACC-06: External tools and extensions

Verification status: Sourced (sourced)
Support: partial

MCP server is exposed to hosts (no MCP client stated), Claude plugin skill, and install helpers; no plugin marketplace or custom tool registry inside MartinLoop.

Availability: generally-available
Delivery: native
Scope: surfaces=@martinloop/mcp server (stdio and HTTP transports), martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- mcp_client: MartinLoop exposes an MCP server for hosts; bounded search found no MCP client for external tools in the pinned packages (the host agent provides its own).
- plugins: A Claude plugin manifest with a martinloop-govern skill ships; no MartinLoop plugin framework.
- custom_tools: Custom adapters via the SDK; no user-defined tool registry for agents.
- instructions_import: Reads AGENTS.md/CLAUDE.md-style repo context only through the host agent; MartinLoop writes `llms.txt` guidance for agents.
- authentication: `martin mcp install/verify-install` write host config for codex, claude, gemini, cursor and vscode; OAuth is not part of it.
- permission_scope: MCP profiles (minimal, diagnostic, full-local, github-review) limit exposed tools.

- mcp_client — Mcp client [unsupported]: MartinLoop exposes an MCP server for hosts; bounded search found no MCP client for external tools in the pinned packages (the host agent provides its own). Sources: martinloop:mcp-tools, martinloop:srch-team, martinloop:rev-srch-mcpclient.
- plugins — Plugins [partial]: A Claude plugin manifest with a martinloop-govern skill ships; no MartinLoop plugin framework. Sources: martinloop:plugin.
- custom_tools — Custom tools [partial]: Custom adapters via the SDK; no user-defined tool registry for agents. Sources: martinloop:readme, martinloop:examples.
- instructions_import — Instructions import [partial]: Reads AGENTS.md/CLAUDE.md-style repo context only through the host agent; MartinLoop writes `llms.txt` guidance for agents. Sources: martinloop:readme.
- authentication — Authentication [partial]: `martin mcp install/verify-install` write host config for codex, claude, gemini, cursor and vscode; OAuth is not part of it. Sources: martinloop:readme, martinloop:cli-ref.
- permission_scope — Permission scope [partial]: MCP profiles (minimal, diagnostic, full-local, github-review) limit exposed tools. Sources: martinloop:readme.
- integration_owner — Integration owner [supported]: MartinLoop provides the MCP server; any MCP client behavior belongs to the host. Sources: martinloop:mcp-tools.

Checked 2026-11-16 (6 days old on 2026-11-22); not separately cross-checked.

Sources: martinloop:readme, martinloop:mcp-tools, martinloop:plugin, martinloop:srch-team, martinloop:examples, martinloop:cli-ref, martinloop:rev-srch-mcpclient

### ACC-07: Client implementation

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Node/TypeScript CLI, MCP server and SDK; no desktop or mobile client. The hosted dashboard is a closed web app whose implementation is not published.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+), Hosted dashboard / Atlas (closed, Pro plan); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- desktop_implementation: No desktop client; a Node.js/TypeScript terminal CLI (package.json bin martin-loop).
- mobile_implementation: No mobile client in the repository.
- hybrids: No Electron/Tauri hybrid code found.
- implementation_evidence: Package manifest and source show a pnpm TypeScript monorepo; the hosted dashboard (a React single-page app at martinloop.com) is closed.

- desktop_implementation — Desktop implementation [unsupported]: No desktop client; a Node.js/TypeScript terminal CLI (package.json bin martin-loop). Sources: martinloop:readme, martinloop:srch-clients.
- mobile_implementation — Mobile implementation [unsupported]: No mobile client in the repository. Sources: martinloop:srch-clients.
- hybrids — Hybrids [unsupported]: No Electron/Tauri hybrid code found. Sources: martinloop:srch-clients.
- implementation_evidence — Implementation evidence [partial]: Package manifest and source show a pnpm TypeScript monorepo; the hosted dashboard (a React single-page app at martinloop.com) is closed. Sources: martinloop:readme, martinloop:pricing.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:readme, martinloop:srch-clients, martinloop:pricing

## Planning and context

### PLAN-01: Intake and external tracker relationship

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Intake is a prompt/objective given to `martin run` plus an optional swarm plan file; no issue import or external tracker integration.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- external_import: No Linear/Jira/GitHub issue import in the pinned packages.
- chat_git_triggers: No chat or Git-event triggers; a CI example runs the CLI from GitHub Actions manually configured.
- authoritative_tracker: No tracker; the JSONL run store is authoritative for runs only.
- stable_ids: Runs have loop IDs and swarm IDs; there are no issue IDs.
- sync: No tracker sync.
- supported_edits: No issue edits exist.
- external_roundtrips: No external tracker relationship.

- local_creation — Local creation [supported]: `martin run "<objective>"` with acceptance criteria (`--accept`) and metadata creates a run record. Sources: martinloop:cli-ref.
- prompt_intake — Prompt intake [supported]: Objective as positional argument or --objective; MCP martin_run/martin_plan. Sources: martinloop:cli-ref, martinloop:mcp-tools.
- external_import — External import [unsupported]: No Linear/Jira/GitHub issue import in the pinned packages. Sources: martinloop:srch-tracker.
- chat_git_triggers — Chat git triggers [unsupported]: No chat or Git-event triggers; a CI example runs the CLI from GitHub Actions manually configured. Sources: martinloop:srch-sched, martinloop:examples.
- authoritative_tracker — Authoritative tracker [unsupported]: No tracker; the JSONL run store is authoritative for runs only. Sources: martinloop:srch-tracker, martinloop:readme.
- stable_ids — Stable ids [partial]: Runs have loop IDs and swarm IDs; there are no issue IDs. Sources: martinloop:cli-ref.
- sync — Sync [unsupported]: No tracker sync. Sources: martinloop:srch-tracker.
- supported_edits — Supported edits [unsupported]: No issue edits exist. Sources: martinloop:srch-tracker.
- external_roundtrips — External roundtrips [unsupported]: No external tracker relationship. Sources: martinloop:srch-tracker.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:readme, martinloop:cli-ref, martinloop:srch-tracker, martinloop:mcp-tools, martinloop:srch-sched, martinloop:examples

### PLAN-02: Durable task plans

Verification status: Sourced (sourced)
Support: partial

A task contract carries objective, acceptance criteria, verifier plan, scope and budget; `preflight` and `martin_plan` preview it. No plan editor, revision history or mid-run amendment.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+), @martinloop/mcp server (stdio and HTTP transports); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- plans: A Governed Run Plan (objective, verifier plan, scope, budget) is shown before work; it is a contract, not an editable plan document.
- editors: No plan editor; flags, config file or SDK object define the contract.
- revisions: No revision history for contracts; run records are append-only per run.
- implementation_links: Receipts link verifier steps, changed files and rollback evidence to the run; no commit/PR link beyond PR summary.
- amendment: Budget and limits stay bound to the original run; reconnecting does not reset them and there is no mid-run edit.

- plans — Plans [partial]: A Governed Run Plan (objective, verifier plan, scope, budget) is shown before work; it is a contract, not an editable plan document. Sources: martinloop:readme.
- acceptance_criteria — Acceptance criteria [supported]: `--accept <criterion>` (repeatable) adds acceptance criteria stored in the task contract. Sources: martinloop:cli-ref, martinloop:readme.
- editors — Editors [unsupported]: No plan editor; flags, config file or SDK object define the contract. Sources: martinloop:cli-ref, martinloop:rev-srch-edit.
- revisions — Revisions [unsupported]: No revision history for contracts; run records are append-only per run. Sources: martinloop:readme, martinloop:rev-srch-edit.
- implementation_links — Implementation links [partial]: Receipts link verifier steps, changed files and rollback evidence to the run; no commit/PR link beyond PR summary. Sources: martinloop:readme, martinloop:prtools.
- amendment — Amendment [unsupported]: Budget and limits stay bound to the original run; reconnecting does not reset them and there is no mid-run edit. Sources: martinloop:changelog, martinloop:rev-srch-edit.
- restart_persistence — Restart persistence [supported]: Contract, verifier, scope, budget and evidence persist and `continue` reuses the same loop ID. Sources: martinloop:changelog.

Checked 2026-11-16 (6 days old on 2026-11-22); not separately cross-checked.

Sources: martinloop:readme, martinloop:cli-ref, martinloop:mcp-tools, martinloop:changelog, martinloop:prtools, martinloop:rev-srch-edit

### PLAN-03: Hierarchy, dependencies and evolving scope

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Swarm plans carry tasks with dependsOn, assigned agents and write scopes; the scheduler starts tasks only when dependencies are verified. No epics, arbitrary hierarchy or task-to-epic growth.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- epics: No epic entity; a swarm has one parent contract and tasks.
- hierarchy: Parent swarm → tasks → child runs and attempts; depth is fixed, not arbitrary.
- state_propagation: Parent outcome comes only from the sealed parent receipt; child success cannot verify the parent.
- task_to_epic: No conversion of a task into an epic.
- identity_retention: No epic growth; task IDs persist within a swarm plan.
- workflow_constraints: Swarm requires a plan file with explicit agent count and concurrency; single-run mode needs no preset.

- epics — Epics [unsupported]: No epic entity; a swarm has one parent contract and tasks. Sources: martinloop:swarm-cli, martinloop:scheduler.
- hierarchy — Hierarchy [partial]: Parent swarm → tasks → child runs and attempts; depth is fixed, not arbitrary. Sources: martinloop:swarm-cli, martinloop:scheduler.
- blockers — Blockers [supported]: Task `dependsOn` lists block readiness; recovery policy can waive dependencies only when allowed. Sources: martinloop:scheduler.
- readiness — Readiness [supported]: Scheduler selects queued/ready tasks whose dependencies are all satisfied. Sources: martinloop:scheduler.
- enforcement — Enforcement [supported]: Dependency gating is enforced in the scheduler before dispatch, not just displayed. Sources: martinloop:scheduler.
- state_propagation — State propagation [partial]: Parent outcome comes only from the sealed parent receipt; child success cannot verify the parent. Sources: martinloop:hosted-handoff, martinloop:scheduler.
- task_to_epic — Task to epic [unsupported]: No conversion of a task into an epic. Sources: martinloop:swarm-cli.
- identity_retention — Identity retention [unsupported]: No epic growth; task IDs persist within a swarm plan. Sources: martinloop:swarm-cli.
- workflow_constraints — Workflow constraints [partial]: Swarm requires a plan file with explicit agent count and concurrency; single-run mode needs no preset. Sources: martinloop:rel090, martinloop:swarm-cli.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:scheduler, martinloop:swarm-cli, martinloop:rel090, martinloop:hosted-handoff

### PLAN-04: Shared context and memory

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Config file defaults, repo AGENTS.md/CLAUDE.md, a Claude plugin skill and an append-only JSONL memory store; no managed cross-harness memory with correction or deletion controls.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- instructions: `martin.config.yaml` defaults and generated llms.txt guidance; repo instruction files are read by the host agent.
- skills: A martinloop-govern skill ships in the Claude plugin.
- project_docs: Repo analyzer reads project files for scope and style signals.
- memory: Append-only JSONL memory of consent, preference, pattern, feedback and budget entries under the runs root.
- scope: Memory is per runs root (local user); no per-project/team scoping beyond workspace and project IDs.
- provenance: Entries distinguish explicit user input from inferred patterns with confidence scores.
- correction: Entries are never overwritten or edited by design.
- deletion: Append-only; no delete command found in the memory store.

- instructions — Instructions [partial]: `martin.config.yaml` defaults and generated llms.txt guidance; repo instruction files are read by the host agent. Sources: martinloop:config, martinloop:readme.
- skills — Skills [partial]: A martinloop-govern skill ships in the Claude plugin. Sources: martinloop:plugin.
- project_docs — Project docs [partial]: Repo analyzer reads project files for scope and style signals. Sources: martinloop:readme.
- memory — Memory [partial]: Append-only JSONL memory of consent, preference, pattern, feedback and budget entries under the runs root. Sources: martinloop:memory.
- scope — Scope [partial]: Memory is per runs root (local user); no per-project/team scoping beyond workspace and project IDs. Sources: martinloop:memory.
- provenance — Provenance [partial]: Entries distinguish explicit user input from inferred patterns with confidence scores. Sources: martinloop:memory.
- correction — Correction [unsupported]: Entries are never overwritten or edited by design. Sources: martinloop:memory.
- deletion — Deletion [unsupported]: Append-only; no delete command found in the memory store. Sources: martinloop:memory.
- freshness — Freshness [partial]: Inferred memories degrade by confidence; no staleness warnings in review. Sources: martinloop:memory.
- takeover_continuity — Takeover continuity [partial]: Run records and receipts survive a new session on the same machine; hosted sync can share receipts. Sources: martinloop:readme, martinloop:cli-ref.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:config, martinloop:memory, martinloop:plugin, martinloop:readme, martinloop:cli-ref

### PLAN-05: Architecture and API awareness

Verification status: Sourced (sourced)
Support: partial

No code index or knowledge retrieval; the product analyzes the repository for scope/style and risk maps and binds agent output to path boundaries. Architecture enforcement is limited to path and policy checks.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+), @martinloop/mcp server (stdio and HTTP transports); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- code_search: No MartinLoop code search tool; the host agent searches code.
- indexing: A repo analyzer and `martin://repo/risk-map` resource summarize risky paths; no persistent index.
- external_knowledge: No external knowledge connectors in the pinned packages.
- multimodal_inputs: Text objectives only; no image input handling found.
- freshness: No index to go stale.
- access_scope: Allow/deny path globs limit what the agent may write.
- api_retrieval: No retrieval API.
- citations: No retrieval citations.

- code_search — Code search [unsupported]: No MartinLoop code search tool; the host agent searches code. Sources: martinloop:readme, martinloop:rev-srch-index.
- indexing — Indexing [partial]: A repo analyzer and `martin://repo/risk-map` resource summarize risky paths; no persistent index. Sources: martinloop:mcp-tools.
- external_knowledge — External knowledge [unsupported]: No external knowledge connectors in the pinned packages. Sources: martinloop:readme, martinloop:rev-srch-index.
- multimodal_inputs — Multimodal inputs [unsupported]: Text objectives only; no image input handling found. Sources: martinloop:readme, martinloop:rev-srch-image.
- freshness — Freshness [unsupported]: No index to go stale. Sources: martinloop:readme, martinloop:rev-srch-index.
- access_scope — Access scope [partial]: Allow/deny path globs limit what the agent may write. Sources: martinloop:secpol, martinloop:cli-ref.
- api_retrieval — Api retrieval [unsupported]: No retrieval API. Sources: martinloop:readme, martinloop:rev-srch-index.
- citations — Citations [unsupported]: No retrieval citations. Sources: martinloop:readme, martinloop:rev-srch-index.
- conflict_detection — Conflict detection [partial]: Scope, test-integrity and dependency checks detect disallowed changes, not architectural conflicts. Sources: martinloop:secpol.
- review_enforcement — Review enforcement [partial]: Policy and verifier gates run after each attempt; no architecture-consistency review. Sources: martinloop:secpol, martinloop:readme.

Checked 2026-11-16 (6 days old on 2026-11-22); not separately cross-checked.

Sources: martinloop:readme, martinloop:secpol, martinloop:mcp-tools, martinloop:cli-ref, martinloop:rev-srch-index, martinloop:rev-srch-image

### PLAN-06: Native issue management and idea backlog

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

No issue or idea backlog; units are runs and swarm plans stored as JSONL records with loop IDs, triage ranking and metadata.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- create: Runs/swarm plans are created, not issues.
- edit: No editing of a stored run or plan record.
- discuss: No discussion threads.
- prioritize: `triage` ranks saved runs by urgency and missing evidence; no human priority field.
- label: `--metadata key=value` attaches metadata to a run record.
- assign: No assignee field; swarm tasks carry assignedAgentId only.
- search: `runs list` lists saved runs; no full-text search.
- filter: Triage and list views rank; no filter DSL found.

- create — Create [partial]: Runs/swarm plans are created, not issues. Sources: martinloop:cli-ref.
- edit — Edit [unsupported]: No editing of a stored run or plan record. Sources: martinloop:readme.
- discuss — Discuss [unsupported]: No discussion threads. Sources: martinloop:srch-team.
- prioritize — Prioritize [partial]: `triage` ranks saved runs by urgency and missing evidence; no human priority field. Sources: martinloop:cli-ref, martinloop:mcp-tools.
- label — Label [partial]: `--metadata key=value` attaches metadata to a run record. Sources: martinloop:cli-ref.
- assign — Assign [unsupported]: No assignee field; swarm tasks carry assignedAgentId only. Sources: martinloop:swarm-cli.
- search — Search [partial]: `runs list` lists saved runs; no full-text search. Sources: martinloop:cli-ref.
- filter — Filter [partial]: Triage and list views rank; no filter DSL found. Sources: martinloop:cli-ref.
- archive — Archive [unsupported]: No archive command. Sources: martinloop:cli-ref.
- reopen — Reopen [partial]: `resume`/`continue` reopens eligible interrupted reversible runs. Sources: martinloop:changelog.
- project_epic_views — Project epic views [unsupported]: Only workspace/project IDs on records; no project view. Sources: martinloop:readme.
- idea_parking — Idea parking [unsupported]: No way to save an idea without a run record. Sources: martinloop:srch-tracker.
- explicit_dispatch — Explicit dispatch [partial]: `swarm plan` then `swarm run` separate planning from execution; runs start immediately. Sources: martinloop:swarm-cli.
- task_session_separation — Task session separation [partial]: Swarm separates tasks, agents, child runs and attempts. Sources: martinloop:swarm-cli.
- metadata — Metadata [partial]: Run records hold objective, verifier, budget, scope, metadata, attempts and costs. Sources: martinloop:readme.
- durable_states — Durable states [supported]: Lifecycle outcomes VERIFIED, STOPPED, NEEDS REVIEW and stop reasons persist in run records. Sources: martinloop:readme.
- external_tracker_requirement — External tracker requirement [supported]: An external tracker is required for any issue management; none is integrated. Sources: martinloop:srch-tracker.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:readme, martinloop:cli-ref, martinloop:srch-tracker, martinloop:srch-team, martinloop:mcp-tools, martinloop:swarm-cli, martinloop:changelog

### PLAN-07: Agent-native issue operations

Verification status: Sourced (sourced)
Support: partial

An agent host can start, inspect, pause, cancel and summarize runs through MCP, but there are no shared issue IDs or fields because no issue tracker exists.

Availability: generally-available
Delivery: native
Scope: surfaces=@martinloop/mcp server (stdio and HTTP transports); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- shared_ids: Loop and swarm IDs are shared between CLI, MCP and receipts; there are no issue IDs.
- create: martin_run starts a governed run; no issue creation.
- update: No tool edits an existing run record.
- claim: No claim operation.
- assign: No assignment operation.
- state_changes: martin_pause, martin_continue and martin_cancel change run control state on a live local run.
- dependencies: No tool creates dependencies; swarm dependsOn is set in a plan file.
- children: No tool creates children outside swarm plans.

- shared_ids — Shared ids [partial]: Loop and swarm IDs are shared between CLI, MCP and receipts; there are no issue IDs. Sources: martinloop:mcp-tools.
- read — Read [supported]: martin_get_run, martin_list_runs, martin_get_attempt and resources read run records. Sources: martinloop:mcp-tools.
- create — Create [partial]: martin_run starts a governed run; no issue creation. Sources: martinloop:mcp-tools.
- update — Update [unsupported]: No tool edits an existing run record. Sources: martinloop:mcp-tools, martinloop:rev-srch-edit.
- claim — Claim [unsupported]: No claim operation. Sources: martinloop:srch-tracker.
- assign — Assign [unsupported]: No assignment operation. Sources: martinloop:srch-tracker.
- state_changes — State changes [partial]: martin_pause, martin_continue and martin_cancel change run control state on a live local run. Sources: martinloop:mcp-tools.
- dependencies — Dependencies [unsupported]: No tool creates dependencies; swarm dependsOn is set in a plan file. Sources: martinloop:swarm-cli.
- children — Children [unsupported]: No tool creates children outside swarm plans. Sources: martinloop:swarm-cli.
- results — Results [supported]: martin_run_dossier, martin_dossier and martin_pr_summary return results and blockers. Sources: martinloop:mcp-tools.
- human_visibility — Human visibility [partial]: CLI dossier shows the same records; no attributed human view. Sources: martinloop:mcp-tools.
- attribution — Attribution [partial]: Control receipts record local_os_user/local_mcp_client principal, explicitly not a human identity. Sources: martinloop:mcp-tools.
- permissions — Permissions [partial]: Run control only over local stdio with a live control-capable executor; remote transports are rejected. Sources: martinloop:mcp-tools.
- conflicts — Conflicts [partial]: Terminal-state precedence and replay protection govern control requests. Sources: martinloop:rel090.

Checked 2026-11-16 (6 days old on 2026-11-22); not separately cross-checked.

Sources: martinloop:mcp-tools, martinloop:srch-tracker, martinloop:swarm-cli, martinloop:rel090, martinloop:rev-srch-edit

### PLAN-08: Living product specifications

Verification status: Sourced (sourced)
Support: partial

No living product specifications; contracts and acceptance criteria are per run.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- persistent_requirements: No cross-task requirements store; bounded search found only per-run contracts.
- human_edits: No spec to edit.
- agent_edits: No spec to edit.
- revisions: No spec revisions.
- approvals: No spec approvals.
- implementation_links: No requirement-to-implementation links.
- later_retrieval: No retrieval of specs in later tasks.
- drift_detection: No spec drift detection.

- persistent_requirements — Persistent requirements [unsupported]: No cross-task requirements store; bounded search found only per-run contracts. Sources: martinloop:readme, martinloop:cli-ref, martinloop:rev-srch-spec.
- human_edits — Human edits [unsupported]: No spec to edit. Sources: martinloop:readme, martinloop:rev-srch-spec.
- agent_edits — Agent edits [unsupported]: No spec to edit. Sources: martinloop:readme, martinloop:rev-srch-spec.
- revisions — Revisions [unsupported]: No spec revisions. Sources: martinloop:readme, martinloop:rev-srch-spec.
- approvals — Approvals [unsupported]: No spec approvals. Sources: martinloop:readme, martinloop:rev-srch-spec.
- implementation_links — Implementation links [unsupported]: No requirement-to-implementation links. Sources: martinloop:readme, martinloop:rev-srch-spec.
- later_retrieval — Later retrieval [unsupported]: No retrieval of specs in later tasks. Sources: martinloop:readme, martinloop:rev-srch-spec.
- drift_detection — Drift detection [unsupported]: No spec drift detection. Sources: martinloop:readme, martinloop:rev-srch-spec.
- review_integration — Review integration [unsupported]: No spec-aware review. Sources: martinloop:readme, martinloop:rev-srch-spec.

Checked 2026-11-16 (6 days old on 2026-11-22); not separately cross-checked.

Sources: martinloop:readme, martinloop:cli-ref, martinloop:rev-srch-spec

## Coordination and automation

### COORD-01: Parallel work

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Swarm mode runs multiple child agents with explicit max concurrency (default simulation 15 agents, demo concurrency 5); single runs are one attempt loop. Status via `swarm status --watch`.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- quotas: maxConcurrency is a positive integer set per swarm; no global quota or plan limit.
- observed_concurrency: No concurrency measurements are published; the catalog does not measure it.

- concurrent_sessions — Concurrent sessions [supported]: Swarm children run concurrently up to the explicit max concurrency; live runs need explicit agents and concurrency. Sources: martinloop:scheduler, martinloop:rel090.
- concurrent_tasks — Concurrent tasks [supported]: Ready tasks start in parallel up to effective concurrency min(requested, parent). Sources: martinloop:scheduler.
- quotas — Quotas [partial]: maxConcurrency is a positive integer set per swarm; no global quota or plan limit. Sources: martinloop:scheduler.
- observed_concurrency — Observed concurrency [unknown]: No concurrency measurements are published; the catalog does not measure it. Sources: martinloop:readme.
- status_visibility — Status visibility [supported]: `swarm status --watch`, `swarm inspect` and `triage` show running, failed and finished work. Sources: martinloop:swarm-cli, martinloop:cli-ref.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:scheduler, martinloop:swarm-cli, martinloop:readme, martinloop:rel090, martinloop:cli-ref

### COORD-02: Ownership and dispatch

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Swarm plans assign tasks to agent IDs with write scopes and durable execution leases prevent concurrent provider children for a run; no humans, queues or routing UI.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- human_assignment: No human assignment.
- claims: Execution leases give one run one provider child; no task claim by agents.
- queues: Tasks sit queued/ready in a scheduler for one swarm; no cross-run queue.
- routing: Recovery policy can reassign a task up to maxReassignmentsPerTask; routing.ts classifies manager vs direct per task.

- human_assignment — Human assignment [unsupported]: No human assignment. Sources: martinloop:srch-tracker.
- agent_assignment — Agent assignment [supported]: Plan tasks carry assignedAgentId and writeScope. Sources: martinloop:swarm-cli.
- claims — Claims [partial]: Execution leases give one run one provider child; no task claim by agents. Sources: martinloop:changelog.
- queues — Queues [partial]: Tasks sit queued/ready in a scheduler for one swarm; no cross-run queue. Sources: martinloop:scheduler.
- routing — Routing [partial]: Recovery policy can reassign a task up to maxReassignmentsPerTask; routing.ts classifies manager vs direct per task. Sources: martinloop:swarm-cli.
- duplicate_prevention — Duplicate prevention [supported]: Durable execution leases prevent concurrent provider children for one run during reconnects. Sources: martinloop:changelog.
- ownership_enforcement — Ownership enforcement [supported]: Leases and write scopes are enforced in code, not labels. Sources: martinloop:changelog, martinloop:swarm-cli.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:swarm-cli, martinloop:changelog, martinloop:scheduler, martinloop:srch-tracker

### COORD-03: Inspectable delegation and communication

Verification status: Sourced (sourced)
Support: partial

MartinLoop supervises worker agents it launches (child runs in a swarm, assigned agent roles) and returns sealed parent receipts; there is no coordinator agent talking to children.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- roles: Plan agents have role fields; parent contract, child runs, global verifier.
- model_harness_effort: Engine/model per run; per-agent harness choice inside a plan is not documented; no effort setting.
- messages: No inter-agent messaging.
- cross_harness: Whether one swarm can mix harnesses is not stated in plan docs or checked code paths.
- activation: `swarm plan/run` and `demo --swarm`; live mode needs explicit agents, concurrency and positive budget.
- followup: Reassignment and recovery within policy; no open follow-up chat to a finished child.

- roles — Roles [partial]: Plan agents have role fields; parent contract, child runs, global verifier. Sources: martinloop:swarm-cli.
- model_harness_effort — Model harness effort [partial]: Engine and model are chosen per run and fixed for every child of a swarm (parent-selected); there is no per-agent harness or model choice inside a plan and no effort setting. Sources: martinloop:models, martinloop:rev-live-engine.
- messages — Messages [unsupported]: No inter-agent messaging. Sources: martinloop:swarm-cli.
- results_return — Results return [supported]: Child results roll into evidence, global verification and a sealed parent receipt. Sources: martinloop:hosted-handoff.
- cross_harness — Cross harness [unsupported]: One swarm uses a single parent-selected engine and model: the live runtime rejects a child whose adapter provider or model differs, so harnesses cannot be mixed inside one swarm. Sources: martinloop:rev-live-engine, martinloop:models.
- orchestration_owner — Orchestration owner [supported]: MartinLoop core owns orchestration; the coding agent still writes code. Sources: martinloop:readme.
- activation — Activation [partial]: `swarm plan/run` and `demo --swarm`; live mode needs explicit agents, concurrency and positive budget. Sources: martinloop:rel090.
- authentication — Authentication [supported]: Uses the host agent login or API key for workers. Sources: martinloop:models.
- live_progress — Live progress [supported]: `swarm status --watch` shows live child state. Sources: martinloop:swarm-cli.
- retained_results — Retained results [supported]: Swarm dossier, verify and share operate on stored runs. Sources: martinloop:swarm-cli.
- followup — Followup [partial]: Reassignment and recovery within policy; no open follow-up chat to a finished child. Sources: martinloop:swarm-cli.
- invocation — Invocation [supported]: CLI commands and MCP tools invoke runs. Sources: martinloop:cli-ref, martinloop:mcp-tools.

Checked 2026-11-16 (6 days old on 2026-11-22); not separately cross-checked.

Sources: martinloop:swarm-cli, martinloop:hosted-handoff, martinloop:rel090, martinloop:models, martinloop:readme, martinloop:cli-ref, martinloop:mcp-tools, martinloop:rev-live-engine

### COORD-04: Unattended triggers

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

No schedules or event triggers in Core; the CLI can be called from CI (example GitHub Actions budget gate) or cron by the user.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- schedules: No scheduler; bounded search for cron/setInterval found nothing.
- events: No event triggers or webhooks.
- recurring_tasks: No recurring tasks.
- completion_conditions: A run ends on VERIFIED/STOPPED/NEEDS REVIEW against budget, iterations and verifier.
- scheduler_location: No scheduler exists.
- overlaps: No scheduling overlap handling.
- retries: Retries are bounded by --max-iterations within a run.
- client_independence: No product-owned background scheduler; a user-run CI job is a manual integration.

- schedules — Schedules [unsupported]: No scheduler; bounded search for cron/setInterval found nothing. Sources: martinloop:srch-sched.
- events — Events [unsupported]: No event triggers or webhooks. Sources: martinloop:srch-sched.
- recurring_tasks — Recurring tasks [unsupported]: No recurring tasks. Sources: martinloop:srch-sched.
- completion_conditions — Completion conditions [partial]: A run ends on VERIFIED/STOPPED/NEEDS REVIEW against budget, iterations and verifier. Sources: martinloop:readme.
- scheduler_location — Scheduler location [unsupported]: No scheduler exists. Sources: martinloop:srch-sched.
- overlaps — Overlaps [unsupported]: No scheduling overlap handling. Sources: martinloop:srch-sched.
- retries — Retries [partial]: Retries are bounded by --max-iterations within a run. Sources: martinloop:cli-ref.
- client_independence — Client independence [unsupported]: No product-owned background scheduler; a user-run CI job is a manual integration. Sources: martinloop:examples.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:examples, martinloop:srch-sched, martinloop:readme, martinloop:cli-ref

### COORD-05: Intervention and attention

Verification status: Sourced (sourced)
Support: partial

Pause/continue/cancel through local stdio MCP with receipts; `swarm cancel` and `resume`/`continue` on the CLI; hard stops preserve history. No notifications or live steering.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+), @martinloop/mcp server (stdio and HTTP transports); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- steer: No live message to the running agent; objective and limits are fixed per run.
- notifications: No notification code.
- pending_decisions: Blocked runs report the exact approval flag to rerun with; no queued decision inbox.
- escalation: Policy failures stop or escalate to NEEDS REVIEW with a next safe action.
- chat_terminal_parity: Control mutations work only over local stdio MCP and CLI; remote/HTTP transports reject them.

- pause — Pause [supported]: martin_pause requests a pause at a safe boundary on a live local run. Sources: martinloop:mcp-tools.
- stop — Stop [supported]: martin_cancel and `swarm cancel` publish durable cancellation and capture a receipt. Sources: martinloop:mcp-tools, martinloop:swarm-cli.
- resume — Resume [supported]: `resume`/`continue <loopId>` and martin_continue continue eligible runs with recorded context. Sources: martinloop:cli-ref, martinloop:changelog.
- steer — Steer [unsupported]: No live message to the running agent; objective and limits are fixed per run. Sources: martinloop:changelog, martinloop:rev-srch-steer.
- notifications — Notifications [unsupported]: No notification code. Sources: martinloop:srch-notify.
- pending_decisions — Pending decisions [partial]: Blocked runs report the exact approval flag to rerun with; no queued decision inbox. Sources: martinloop:leash.
- escalation — Escalation [partial]: Policy failures stop or escalate to NEEDS REVIEW with a next safe action. Sources: martinloop:readme, martinloop:secpol.
- visible_errors — Visible errors [supported]: Failure class (13), stop reason and verifier output appear in the dossier. Sources: martinloop:readme.
- chat_terminal_parity — Chat terminal parity [partial]: Control mutations work only over local stdio MCP and CLI; remote/HTTP transports reject them. Sources: martinloop:mcp-tools.
- retained_history — Retained history [supported]: Stopped and cancelled runs keep records and receipts. Sources: martinloop:rel090.

Checked 2026-11-16 (6 days old on 2026-11-22); not separately cross-checked.

Sources: martinloop:mcp-tools, martinloop:swarm-cli, martinloop:rel090, martinloop:changelog, martinloop:srch-notify, martinloop:cli-ref, martinloop:leash, martinloop:readme, martinloop:secpol, martinloop:rev-srch-steer

### COORD-06: Fleet and epic management UI

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

No fleet or epic UI in Core; `triage`, `runs list` and swarm status/inspect give terminal views, and a closed hosted dashboard lists loops.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+), Hosted dashboard / Atlas (closed, Pro plan); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- fleet_fixture: Swarm demo shows 15 agents on a launch board in the terminal; the hosted fleet UI is closed.
- parent_child: Swarm inspect shows parent, tasks, children and attempts.
- dependencies: Task dependsOn is part of the swarm plan view.
- attention: Triage ranks runs by urgency and missing evidence.
- workload: Status shows running count against concurrency; no workload dashboard.
- navigation_errors: Failure class and stop reason shown per run.

- fleet_fixture — Fleet fixture [partial]: Swarm demo shows 15 agents on a launch board in the terminal; the hosted fleet UI is closed. Sources: martinloop:readme, martinloop:pricing.
- parent_child — Parent child [partial]: Swarm inspect shows parent, tasks, children and attempts. Sources: martinloop:swarm-cli.
- dependencies — Dependencies [partial]: Task dependsOn is part of the swarm plan view. Sources: martinloop:scheduler.
- attention — Attention [partial]: Triage ranks runs by urgency and missing evidence. Sources: martinloop:cli-ref.
- history — History [supported]: Run history and dossiers persist. Sources: martinloop:readme.
- workload — Workload [partial]: Status shows running count against concurrency; no workload dashboard. Sources: martinloop:swarm-cli.
- navigation_errors — Navigation errors [partial]: Failure class and stop reason shown per run. Sources: martinloop:readme.
- fixture_mode — Fixture mode [supported]: `demo --swarm` is a deterministic provider-free simulation labelled non-governed. Sources: martinloop:readme.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:swarm-cli, martinloop:cli-ref, martinloop:pricing, martinloop:readme, martinloop:scheduler

### COORD-07: Workflow flexibility and orchestration visibility

Verification status: Sourced (sourced)
Support: partial

Single governed run or swarm plan; no presets or natural-language configuration, and settings are fixed per run. Orchestration is visible through swarm inspect and receipts.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- growing_epic: No growth from task to epic.
- presets: `policyProfile` (e.g. balanced) and execution profiles (strict_local, ci_safe, staging_controlled, research_untrusted).
- natural_language: Configuration is flags, YAML and JSON plan files.
- agent_configuration: `--engine`/`--model` per run; swarm agents carry contracts in the plan.
- in_run_changes: Limits and scope stay bound to the original run.
- auth_friction: Needs the host agent CLI login or an API key; hosted sync needs a token.

- quick_task — Quick task [supported]: `martin run "<task>" --verify ...` runs one task with no preset. Sources: martinloop:readme.
- growing_epic — Growing epic [unsupported]: No growth from task to epic. Sources: martinloop:swarm-cli.
- presets — Presets [partial]: `policyProfile` (e.g. balanced) and execution profiles (strict_local, ci_safe, staging_controlled, research_untrusted). Sources: martinloop:config, martinloop:leash.
- natural_language — Natural language [unsupported]: Configuration is flags, YAML and JSON plan files. Sources: martinloop:config.
- agent_configuration — Agent configuration [partial]: `--engine`/`--model` per run; swarm agents carry contracts in the plan. Sources: martinloop:models, martinloop:swarm-cli.
- in_run_changes — In run changes [unsupported]: Limits and scope stay bound to the original run. Sources: martinloop:changelog, martinloop:rev-srch-edit.
- live_roles — Live roles [supported]: Swarm status shows child agents, roles and statuses live. Sources: martinloop:swarm-cli.
- retained_relationships — Retained relationships [supported]: Receipts and evidence index keep parent/child/attempt relations. Sources: martinloop:hosted-handoff.
- discoverability — Discoverability [supported]: `start`, `doctor`, `tour` and docs guide new users; MCP prompts expose workflows. Sources: martinloop:cli-ref, martinloop:mcp-tools.
- auth_friction — Auth friction [partial]: Needs the host agent CLI login or an API key; hosted sync needs a token. Sources: martinloop:models, martinloop:cli-ref.

Checked 2026-11-16 (6 days old on 2026-11-22); not separately cross-checked.

Sources: martinloop:readme, martinloop:swarm-cli, martinloop:config, martinloop:changelog, martinloop:leash, martinloop:models, martinloop:hosted-handoff, martinloop:cli-ref, martinloop:mcp-tools, martinloop:rev-srch-edit

## Execution and reliability

### RUN-01: Laptop-independent execution

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Runs execute on the user's machine (the agent CLI process on the local checkout). No BYO-remote, self-hosted service, managed cloud or hybrid execution; hosted sync only uploads receipts.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+), Hosted dashboard / Atlas (closed, Pro plan); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- byo_remote: No remote host launcher in the pinned packages.
- self_hosted: No self-hosted execution service; Enterprise VPC/on-prem is a sales offering whose code is not public.
- managed_cloud: No managed-cloud runner; hosted Pro is a dashboard of receipts.
- hybrid: Local execution with optional hosted receipt sync (`sync flush`).
- provisioning: No compute provisioning.
- resources: Uses host machine resources; no compute limits.
- region: Hosted region/residency is not published; the hosted service is closed.
- expiry: No remote sessions to expire; local runs persist in the runs root.

- local — Local [supported]: Governed runs execute locally with the agent CLI against the local repository. Sources: martinloop:readme, martinloop:secpol.
- byo_remote — Byo remote [unsupported]: No remote host launcher in the pinned packages. Sources: martinloop:srch-isolation.
- self_hosted — Self hosted [unsupported]: No self-hosted execution service; Enterprise VPC/on-prem is a sales offering whose code is not public. Sources: martinloop:pricing, martinloop:srch-isolation.
- managed_cloud — Managed cloud [unsupported]: No managed-cloud runner; hosted Pro is a dashboard of receipts. Sources: martinloop:pricing, martinloop:hosted-handoff.
- hybrid — Hybrid [partial]: Local execution with optional hosted receipt sync (`sync flush`). Sources: martinloop:cli-ref.
- provisioning — Provisioning [unsupported]: No compute provisioning. Sources: martinloop:srch-isolation.
- payer — Payer [supported]: User pays the agent provider; hosted Pro is $49/month. Sources: martinloop:pricing, martinloop:models.
- resources — Resources [unsupported]: Uses host machine resources; no compute limits. Sources: martinloop:readme.
- region — Region [unknown]: Hosted region/residency is not published; the hosted service is closed. Sources: martinloop:pricing.
- expiry — Expiry [unsupported]: No remote sessions to expire; local runs persist in the runs root. Sources: martinloop:readme.
- client_disconnect — Client disconnect [partial]: Closing the terminal ends the local process; `continue` can resume eligible interrupted reversible runs. Sources: martinloop:changelog.
- other_device_continuation — Other device continuation [unsupported]: Execution cannot move to another device. Sources: martinloop:hosted-handoff.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:readme, martinloop:secpol, martinloop:hosted-handoff, martinloop:srch-isolation, martinloop:pricing, martinloop:cli-ref, martinloop:models, martinloop:changelog

### RUN-02: Workspace and process isolation

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

No sandbox for the agent process; MartinLoop observes repository changes with allow/deny path rules, secret-file rules and rollback. Swarm children use Git worktrees (or independent clones) at a baseline commit.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- branch: Compares changes with the commit checked out at attempt start; no branch creation.
- container: No container launcher; bounded search found only docker patterns for command blocking.
- vm: No VM isolation.
- os_sandbox: Release notes state it is not an operating-system sandbox; Codex/Claude sandboxes belong to the provider.
- filesystem: Allow/deny globs, symlink and junction collapse, secret-file protection and rollback check repo writes; writes outside the repo are not contained.
- process: Agent process is not contained or limited.
- network: `--allow-network-domain` with execution profiles governs declared network authority for staging; agent network access is the provider's.
- secrets: Writes to .env/key files are blocked unless explicitly allowed; secret-like task text is refused.

- checkout — Checkout [supported]: Single governed runs act in the configured repository checkout (--cwd). Sources: martinloop:secpol, martinloop:cli-ref.
- branch — Branch [partial]: Compares changes with the commit checked out at attempt start; no branch creation. Sources: martinloop:secpol.
- worktree — Worktree [supported]: Swarm child, integration and verifier workspaces are Git worktrees (or independent clones) at a baseline commit. Sources: martinloop:workspaces.
- container — Container [unsupported]: No container launcher; bounded search found only docker patterns for command blocking. Sources: martinloop:srch-isolation.
- vm — Vm [unsupported]: No VM isolation. Sources: martinloop:srch-isolation.
- os_sandbox — Os sandbox [unsupported]: Release notes state it is not an operating-system sandbox; Codex/Claude sandboxes belong to the provider. Sources: martinloop:rel090, martinloop:secpol.
- filesystem — Filesystem [partial]: Allow/deny globs, symlink and junction collapse, secret-file protection and rollback check repo writes; writes outside the repo are not contained. Sources: martinloop:secpol.
- process — Process [unsupported]: Agent process is not contained or limited. Sources: martinloop:secpol.
- network — Network [partial]: `--allow-network-domain` with execution profiles governs declared network authority for staging; agent network access is the provider's. Sources: martinloop:leash, martinloop:extout.
- secrets — Secrets [partial]: Writes to .env/key files are blocked unless explicitly allowed; secret-like task text is refused. Sources: martinloop:secpol.
- code_separation — Code separation [partial]: Swarm children are separated in worktrees; single runs share one checkout. Sources: martinloop:workspaces.
- security_boundary — Security boundary [unsupported]: No hostile-code containment claimed; use an isolated sandbox for host-level write prevention. Sources: martinloop:rel090.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:secpol, martinloop:rel090, martinloop:workspaces, martinloop:leash, martinloop:srch-isolation, martinloop:cli-ref, martinloop:extout

### RUN-03: Recovery, expiry and continuity

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Durable JSONL records, execution leases, retained rollback boundaries and `continue` support reconnect after interruption for reversible runs; terminal/exhausted/irreversible runs fail closed rather than replay.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- client_disconnect: A closed terminal interrupts the local process; eligible interrupted runs can `continue`.
- worker_crash: Workspace-observation failure ends the attempt as a typed failure with safe rollback; stale-lock recovery is hardened.
- service_restart: Durable execution leases and control journals survive reconnects (PID reuse hardened).
- expiry: No sandbox expiry exists.
- process_reconnect: Reconnect reuses the same loop and one provider child via leases.
- state_reconstruction: Retained candidate rollback authority and latest verifier failure survive reconnects.
- context_loss: Provider session context is not stated to be retained; MartinLoop retains its own evidence.
- manual_repair: Quarantine recovery stays explicit when the workspace cannot be restored safely.

- client_disconnect — Client disconnect [partial]: A closed terminal interrupts the local process; eligible interrupted runs can `continue`. Sources: martinloop:changelog.
- worker_crash — Worker crash [partial]: Workspace-observation failure ends the attempt as a typed failure with safe rollback; stale-lock recovery is hardened. Sources: martinloop:changelog, martinloop:rel090.
- service_restart — Service restart [partial]: Durable execution leases and control journals survive reconnects (PID reuse hardened). Sources: martinloop:rel090.
- expiry — Expiry [unsupported]: No sandbox expiry exists. Sources: martinloop:srch-isolation.
- transcript_reload — Transcript reload [supported]: `dossier`, `runs get` and `runs attempt` reload persisted records. Sources: martinloop:cli-ref.
- process_reconnect — Process reconnect [partial]: Reconnect reuses the same loop and one provider child via leases. Sources: martinloop:changelog.
- task_resume — Task resume [supported]: `continue <loopId>` reuses persisted task, verifier, scope, budget and evidence for eligible runs. Sources: martinloop:changelog.
- state_reconstruction — State reconstruction [partial]: Retained candidate rollback authority and latest verifier failure survive reconnects. Sources: martinloop:changelog.
- context_loss — Context loss [partial]: Provider session context is not stated to be retained; MartinLoop retains its own evidence. Sources: martinloop:changelog.
- duplicate_actions — Duplicate actions [supported]: Terminal, exhausted, uncertain-active, irreversible and external-write runs are not replayed automatically. Sources: martinloop:changelog.
- manual_repair — Manual repair [partial]: Quarantine recovery stays explicit when the workspace cannot be restored safely. Sources: martinloop:rel090.
- multi_day_continuity — Multi day continuity [partial]: Records persist locally and limits stay bound across reconnects; no compute replacement exists. Sources: martinloop:changelog.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:changelog, martinloop:rel090, martinloop:rollback, martinloop:workspaces, martinloop:srch-isolation, martinloop:cli-ref

### RUN-04: Development environment reliability

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

No environment management: dependencies, caches, services and ports are the user's; rollback ignores dependency/build directories. Swarm worktrees are created from a baseline commit.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- dependencies: No dependency installation step; the demo asks the user to run npm install.
- caches: Rollback skips file-by-file inspection of node_modules, dist and coverage directories.
- snapshots: Rollback boundary snapshots of repository files; no environment snapshots.
- services: No service management.
- ports: No port allocation.
- secrets: Secret-file write blocking only; no secret injection.
- duplicate_installs: Swarm worktrees have no shared install handling in the pinned code.
- collisions: No port/service collision handling.

- dependencies — Dependencies [unsupported]: No dependency installation step; the demo asks the user to run npm install. Sources: martinloop:readme, martinloop:srch-isolation.
- caches — Caches [partial]: Rollback skips file-by-file inspection of node_modules, dist and coverage directories. Sources: martinloop:secpol.
- snapshots — Snapshots [partial]: Rollback boundary snapshots of repository files; no environment snapshots. Sources: martinloop:rollback.
- services — Services [unsupported]: No service management. Sources: martinloop:srch-isolation.
- ports — Ports [unsupported]: No port allocation. Sources: martinloop:srch-isolation.
- secrets — Secrets [partial]: Secret-file write blocking only; no secret injection. Sources: martinloop:secpol.
- duplicate_installs — Duplicate installs [unsupported]: Swarm worktrees have no shared install handling in the pinned code. Sources: martinloop:workspaces.
- collisions — Collisions [unsupported]: No port/service collision handling. Sources: martinloop:srch-isolation.
- remote_setup — Remote setup [unsupported]: No remote environments. Sources: martinloop:srch-isolation.
- automatic_repair — Automatic repair [partial]: `doctor` diagnoses agent readiness; cleanup recovery handles workspaces after failures. Sources: martinloop:cli-ref, martinloop:rel090.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:secpol, martinloop:workspaces, martinloop:srch-isolation, martinloop:readme, martinloop:rollback, martinloop:cli-ref, martinloop:rel090

### RUN-05: Limits and failure handling

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Hard USD, token, iteration, provider-timeout and verifier-timeout limits enforced before the next attempt; soft limit warns; provider fallback is explicit.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- time_limits: Provider execution and verifier timeouts exist in request options; USD/iteration limits are the headline controls.
- timeouts: verifyTimeoutMs and providerExecutionTimeoutMs are request fields.
- provider_failures: Failures are classified into 13 runtime classes and may trigger the configured fallback engine; external-write recovery stays disabled.

- time_limits — Time limits [partial]: Provider execution and verifier timeouts exist in request options; USD/iteration limits are the headline controls. Sources: martinloop:readme.
- turn_limits — Turn limits [supported]: `--max-iterations` caps attempts. Sources: martinloop:cli-ref.
- token_limits — Token limits [supported]: `--max-tokens` and budget.maxTokens. Sources: martinloop:cli-ref, martinloop:config.
- spend_limits — Spend limits [supported]: `--budget-usd` hard cap and `--soft-limit-usd` warning; budget preflight blocks attempts that would exceed policy. Sources: martinloop:cli-ref, martinloop:readme.
- timeouts — Timeouts [partial]: verifyTimeoutMs and providerExecutionTimeoutMs are request fields. Sources: martinloop:cli-ref.
- retries — Retries [supported]: Bounded by max iterations; provider fallback chain is explicit and validated before spend. Sources: martinloop:changelog.
- provider_failures — Provider failures [partial]: Failures are classified into 13 runtime classes and may trigger the configured fallback engine; external-write recovery stays disabled. Sources: martinloop:readme, martinloop:changelog.
- enforcement — Enforcement [supported]: Budget caps stop the next attempt (budget_exit); limits cannot reset on reconnect. Sources: martinloop:readme, martinloop:changelog.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:readme, martinloop:cli-ref, martinloop:config, martinloop:changelog

### RUN-06: Truthful status and operational statistics

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

JSONL run records, per-attempt receipts, verifier evidence, 13-class failure classification, HMAC integrity verdicts, triage and dossier views. No benchmark performance numbers are produced by this catalog.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+), @martinloop/mcp server (stdio and HTTP transports); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- usage_export: `share` writes a redacted JSON/Markdown/SVG bundle; `audit --json` for Claude Code history; no CSV usage export.
- task_stats: Per-run cost, attempts and verifier results in the dossier.
- agent_stats: Runs record runtime/engine; no per-agent statistics view found.
- project_stats: Records carry workspace/project IDs; project aggregation is a hosted feature (closed).
- time_stats: Audit reports retry chains and cost over a day range; run durations in records.
- approval_status: Approval flags and policy blocks appear in the failure reason.
- performance_method: A benchmark workspace exists but the catalog does not run it and MartinLoop is not on ADEMark.
- responsiveness: No responsiveness measurements published.

- logs — Logs [supported]: Run records, `martin_logs` receipts and workflow events. Sources: martinloop:mcp-tools, martinloop:readme.
- traces — Traces [supported]: Per-attempt records, event chain and trace store with receipt integrity. Sources: martinloop:readme.
- failure_reasons — Failure reasons [supported]: 13 runtime failure classes and stop reasons. Sources: martinloop:readme.
- usage_export — Usage export [partial]: `share` writes a redacted JSON/Markdown/SVG bundle; `audit --json` for Claude Code history; no CSV usage export. Sources: martinloop:cli-ref, martinloop:readme.
- task_stats — Task stats [partial]: Per-run cost, attempts and verifier results in the dossier. Sources: martinloop:readme.
- agent_stats — Agent stats [partial]: Runs record runtime/engine; no per-agent statistics view found. Sources: martinloop:readme.
- project_stats — Project stats [partial]: Records carry workspace/project IDs; project aggregation is a hosted feature (closed). Sources: martinloop:cli-ref, martinloop:pricing.
- time_stats — Time stats [partial]: Audit reports retry chains and cost over a day range; run durations in records. Sources: martinloop:readme.
- failure_status — Failure status [supported]: Outcomes VERIFIED, STOPPED, NEEDS REVIEW with failure class. Sources: martinloop:readme.
- approval_status — Approval status [partial]: Approval flags and policy blocks appear in the failure reason. Sources: martinloop:leash.
- reconnect_consistency — Reconnect consistency [supported]: Terminal-state precedence and journals keep state consistent after reconnect. Sources: martinloop:rel090.
- performance_method — Performance method [unknown]: A benchmark workspace exists but the catalog does not run it and MartinLoop is not on ADEMark. Sources: martinloop:readme.
- responsiveness — Responsiveness [unknown]: No responsiveness measurements published. Sources: martinloop:readme.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:readme, martinloop:cli-ref, martinloop:mcp-tools, martinloop:pricing, martinloop:leash, martinloop:rel090

### RUN-07: Per-task API and compute cost

Verification status: Sourced (sourced)
Support: partial

Per-run and per-attempt cost and tokens with provenance labels actual/calculated/estimated/unavailable; Codex usage is estimated unless the host exposes it; `audit` estimates API-equivalent spend in Claude Code history. No compute cost.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- child_cost: Swarm settlement reconciles child spend with the run ledger for live runs.
- epic_cost: No epic entity; swarm total only.
- harness_aggregation: Fallback chain keeps one run record; cross-harness aggregation beyond one run is not shown.
- subscription_allowance: `audit` reports API-equivalent spend for subscription users; actual subscription marginal cost is not computed.
- provider_reconciliation: Live Swarm requires settlement evidence that reconciles with the run ledger.
- compute_cost: No compute fees exist.
- refresh_delay: Usage refresh timing depends on the provider host and is not documented.
- alerts: Soft limit threshold warns before the hard cap.

- task_cost — Task cost [supported]: Each run has cost_usd with provenance in the proof receipt/dossier. Sources: martinloop:readme.
- session_cost — Session cost [supported]: Run and attempt level cost in dossiers. Sources: martinloop:readme.
- child_cost — Child cost [partial]: Swarm settlement reconciles child spend with the run ledger for live runs. Sources: martinloop:rel090.
- epic_cost — Epic cost [unsupported]: No epic entity; swarm total only. Sources: martinloop:rel090, martinloop:rev-srch-epic.
- harness_aggregation — Harness aggregation [partial]: Fallback chain keeps one run record; cross-harness aggregation beyond one run is not shown. Sources: martinloop:changelog.
- billed_vs_estimated — Billed vs estimated [supported]: Provenance labels actual, calculated, estimated, unavailable are always shown. Sources: martinloop:readme.
- tokens — Tokens [supported]: Token counts and --max-tokens budget. Sources: martinloop:readme, martinloop:cli-ref.
- subscription_allowance — Subscription allowance [partial]: `audit` reports API-equivalent spend for subscription users; actual subscription marginal cost is not computed. Sources: martinloop:readme, martinloop:cli-ref.
- provider_reconciliation — Provider reconciliation [partial]: Live Swarm requires settlement evidence that reconciles with the run ledger. Sources: martinloop:rel090.
- compute_cost — Compute cost [unsupported]: No compute fees exist. Sources: martinloop:readme, martinloop:rev-srch-cost.
- refresh_delay — Refresh delay [unknown]: Usage refresh timing depends on the provider host and is not documented. Sources: martinloop:readme.
- alerts — Alerts [partial]: Soft limit threshold warns before the hard cap. Sources: martinloop:cli-ref.
- export — Export [partial]: `share --latest` bundle and `dossier --json`. Sources: martinloop:cli-ref.
- drilldown — Drilldown [supported]: Dossier → attempts via `runs attempt`. Sources: martinloop:cli-ref.

Checked 2026-11-16 (6 days old on 2026-11-22); not separately cross-checked.

Sources: martinloop:readme, martinloop:cli-ref, martinloop:rel090, martinloop:changelog, martinloop:rev-srch-cost, martinloop:rev-srch-epic

### RUN-08: Compute-aware scheduling and verification

Verification status: Sourced (sourced)
Support: partial

Swarm scheduler enforces explicit maximum concurrency and dependency gating; sibling children get fair-share budget lease reservations. No CPU/RAM awareness, host awareness or check deduplication.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- cpu_ram: No CPU/RAM limits found in scheduler.
- backpressure: Excess ready tasks wait for slots; no resource backpressure signal.
- host_awareness: No per-host resource awareness.
- check_reuse: No check reuse found.
- check_deduplication: No check deduplication found.
- bounded_tests: Verifier timeouts and external-outcome deadline with four concurrent observations.
- fairness: Fair-share reservations are kept for sibling tasks.
- recovery: Recovery policy reassigns tasks within maxReassignmentsPerTask.

- concurrency — Concurrency [supported]: maxConcurrency must be a positive integer; effective = min(requested, parent maxConcurrency); live runs require it. Sources: martinloop:scheduler, martinloop:rel090.
- queues — Queues [supported]: Tasks wait in queued/ready states until a slot and dependencies allow. Sources: martinloop:scheduler.
- cpu_ram — Cpu ram [unsupported]: No CPU/RAM limits found in scheduler. Sources: martinloop:scheduler.
- backpressure — Backpressure [partial]: Excess ready tasks wait for slots; no resource backpressure signal. Sources: martinloop:scheduler.
- host_awareness — Host awareness [unsupported]: No per-host resource awareness. Sources: martinloop:scheduler.
- check_reuse — Check reuse [unsupported]: No check reuse found. Sources: martinloop:scheduler.
- check_deduplication — Check deduplication [unsupported]: No check deduplication found. Sources: martinloop:scheduler.
- bounded_tests — Bounded tests [partial]: Verifier timeouts and external-outcome deadline with four concurrent observations. Sources: martinloop:extout.
- cancellation — Cancellation [supported]: `swarm cancel` and martin_cancel. Sources: martinloop:swarm-cli, martinloop:mcp-tools.
- fairness — Fairness [partial]: Budget leases for sibling children are reserved fair-share and reclaimed only as needed; no fairness policy for task or slot ordering is documented. Sources: martinloop:scheduler, martinloop:rev-sched-fair.
- recovery — Recovery [partial]: Recovery policy reassigns tasks within maxReassignmentsPerTask. Sources: martinloop:swarm-cli.
- shared_limits — Shared limits [unsupported]: One swarm has its own limits; no cross-swarm shared limits. Sources: martinloop:scheduler.

Checked 2026-11-16 (6 days old on 2026-11-22); not separately cross-checked.

Sources: martinloop:scheduler, martinloop:rel090, martinloop:swarm-cli, martinloop:extout, martinloop:mcp-tools, martinloop:rev-sched-fair

## Human collaboration and control

### TEAM-01: Live human collaboration

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Single-user local tool; shared dashboards (Growth) are waitlist only. A share bundle can be handed to another person but there is no shared visibility, presence or co-prompting.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+), Hosted dashboard / Atlas (closed, Pro plan); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- visibility: `share` produces a redacted receipt bundle for another person; the hosted Pro dashboard is personal-project only; Growth shared dashboards are waitlist.
- discussion: No discussion code.
- comments: No comments.
- presence: No presence.
- coprompting: No co-prompting; one local operator.
- ordering: No multi-user input.
- conflicts: No multi-user conflicts handled.
- attribution: Control receipts record a local OS user principal only.

- visibility — Visibility [partial]: `share` produces a redacted receipt bundle for another person; the hosted Pro dashboard is personal-project only; Growth shared dashboards are waitlist. Sources: martinloop:readme, martinloop:pricing.
- discussion — Discussion [unsupported]: No discussion code. Sources: martinloop:srch-team.
- comments — Comments [unsupported]: No comments. Sources: martinloop:srch-team.
- presence — Presence [unsupported]: No presence. Sources: martinloop:srch-team.
- coprompting — Coprompting [unsupported]: No co-prompting; one local operator. Sources: martinloop:srch-team.
- ordering — Ordering [unsupported]: No multi-user input. Sources: martinloop:srch-team.
- conflicts — Conflicts [unsupported]: No multi-user conflicts handled. Sources: martinloop:srch-team.
- attribution — Attribution [partial]: Control receipts record a local OS user principal only. Sources: martinloop:mcp-tools.
- permissions — Permissions [unsupported]: No sharing permissions. Sources: martinloop:srch-team.
- plan_restrictions — Plan restrictions [partial]: Pro is personal projects; shared team dashboard is Growth (waitlist). Sources: martinloop:pricing.
- owner_online — Owner online [unsupported]: Not applicable to live collaboration; local runs need the owner machine. Sources: martinloop:readme.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:readme, martinloop:pricing, martinloop:srch-team, martinloop:mcp-tools

### TEAM-02: Roles and access boundaries

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

No roles in Core; Enterprise pricing card lists RBAC + SSO and audit-log retention, but it is contact-sales with no public documentation.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+), Hosted dashboard / Atlas (closed, Pro plan); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- organization_roles: Enterprise card lists RBAC, but the hosted service is closed and unreleased publicly.
- project_roles: Not documented; closed hosted service.
- task_roles: No task roles in Core.
- membership: Membership lifecycle for the hosted product is not documented.
- sso: Enterprise card lists SSO as contact-sales; not documented or testable.
- provisioning: No provisioning documentation.
- unauthorized_access: Local files and MCP control rely on the OS user; hosted sync needs a workspace token.
- revocation: Hosted token revocation is not documented.

- organization_roles — Organization roles [unknown]: Enterprise card lists RBAC, but the hosted service is closed and unreleased publicly. Sources: martinloop:pricing.
- project_roles — Project roles [unknown]: Not documented; closed hosted service. Sources: martinloop:pricing.
- task_roles — Task roles [unsupported]: No task roles in Core. Sources: martinloop:srch-team.
- membership — Membership [unknown]: Membership lifecycle for the hosted product is not documented. Sources: martinloop:pricing.
- sso — Sso [unknown]: Enterprise card lists SSO as contact-sales; not documented or testable. Sources: martinloop:pricing.
- provisioning — Provisioning [unknown]: No provisioning documentation. Sources: martinloop:pricing.
- unauthorized_access — Unauthorized access [partial]: Local files and MCP control rely on the OS user; hosted sync needs a workspace token. Sources: martinloop:mcp-tools, martinloop:cli-ref.
- revocation — Revocation [unknown]: Hosted token revocation is not documented. Sources: martinloop:cli-ref.
- plan_scope — Plan scope [partial]: Enterprise only (contact sales); Free/Pro have no roles. Sources: martinloop:pricing.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:pricing, martinloop:srch-team, martinloop:mcp-tools, martinloop:cli-ref

### TEAM-03: Approvals and temporary access

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Approvals are explicit pre-run flags (`--approve-test-changes`, dependency, migration, config, external writes) enforced by a policy leash with allow/deny path lists; no interactive approval queue.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- command: Verifier commands and destructive commands are checked and blocked; `destructiveActionPolicy: approval` in config.
- tool: MCP tool profiles limit exposed tools; run control is stdio-only.
- network: `--allow-network-domain` with execution profiles for staging runs.
- delivery: `martin_create_pr` requires a VERIFIED run with verified receipt and `execute=true`.
- approver_identity: Flags carry no approver identity; principal is the local OS user.
- lifetime: Approvals apply to the single invocation that carries the flag.
- revoke: Cancel a live run; no standing approvals to revoke.
- tunnels: No tunnels.

- command — Command [partial]: Verifier commands and destructive commands are checked and blocked; `destructiveActionPolicy: approval` in config. Sources: martinloop:leash, martinloop:config.
- tool — Tool [partial]: MCP tool profiles limit exposed tools; run control is stdio-only. Sources: martinloop:readme, martinloop:mcp-tools.
- network — Network [partial]: `--allow-network-domain` with execution profiles for staging runs. Sources: martinloop:extout, martinloop:leash.
- secret — Secret [supported]: Secret-file writes blocked unless the file is named in --allow-path. Sources: martinloop:secpol.
- delivery — Delivery [partial]: `martin_create_pr` requires a VERIFIED run with verified receipt and `execute=true`. Sources: martinloop:prtools.
- approver_identity — Approver identity [unsupported]: Flags carry no approver identity; principal is the local OS user. Sources: martinloop:mcp-tools.
- lifetime — Lifetime [partial]: Approvals apply to the single invocation that carries the flag. Sources: martinloop:cli-ref.
- allow — Allow [supported]: `--allow-path`, `--approve-*` flags. Sources: martinloop:cli-ref.
- deny — Deny [supported]: `--deny-path` and policy blocks. Sources: martinloop:cli-ref, martinloop:secpol.
- revoke — Revoke [partial]: Cancel a live run; no standing approvals to revoke. Sources: martinloop:mcp-tools.
- tunnels — Tunnels [unsupported]: No tunnels. Sources: martinloop:srch-isolation.
- defaults — Defaults [partial]: Default policy profile balanced; `--unsafe-allow-unguarded-run` bypasses the gate for one run. Sources: martinloop:config, martinloop:cli-ref.
- safer_modes — Safer modes [supported]: Execution profiles strict_local, ci_safe, staging_controlled, research_untrusted. Sources: martinloop:leash.
- enforcement — Enforcement [supported]: Enforced in product code before execution, not by prompt. Sources: martinloop:leash, martinloop:secpol.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:leash, martinloop:secpol, martinloop:config, martinloop:cli-ref, martinloop:readme, martinloop:mcp-tools, martinloop:extout, martinloop:prtools, martinloop:srch-isolation

### TEAM-04: Decision and activity history

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Append-only run records with HMAC integrity (verified / tamper_detected / unsigned) preserve attempts, verifier results and control receipts; single-user, no export beyond share bundles, and the HMAC key is local.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- approvals: Approval flags used appear in the run contract/violations; no approver identity.
- edits: Changed files and rollback evidence recorded per attempt.
- discussion: No discussion.
- decisions: Final outcome, stop reason and next safe action recorded.
- cross_user_persistence: Single local user; hosted sync adds a workspace view only on paid plans.
- mutability: Records are locally signed so edits are detectable, but a local user can alter files.
- retention: Hosted Pro lists 30-day log retention; local retention is the user's.
- integrity: HMAC per-runs-root key detects tampering; integrity is evidence, not a universal guarantee.

- actions — Actions [supported]: Attempts, verifier steps, stops and control requests are recorded. Sources: martinloop:readme, martinloop:mcp-tools.
- approvals — Approvals [partial]: Approval flags used appear in the run contract/violations; no approver identity. Sources: martinloop:leash.
- edits — Edits [partial]: Changed files and rollback evidence recorded per attempt. Sources: martinloop:readme.
- discussion — Discussion [unsupported]: No discussion. Sources: martinloop:srch-team.
- decisions — Decisions [partial]: Final outcome, stop reason and next safe action recorded. Sources: martinloop:readme.
- cross_user_persistence — Cross user persistence [unsupported]: Single local user; hosted sync adds a workspace view only on paid plans. Sources: martinloop:pricing.
- cross_session_persistence — Cross session persistence [supported]: Runs persist under ~/.martin/runs. Sources: martinloop:readme.
- export — Export [supported]: `share` bundle (redacted JSON, Markdown, SVG) and `dossier --json`. Sources: martinloop:readme, martinloop:cli-ref.
- mutability — Mutability [partial]: Records are locally signed so edits are detectable, but a local user can alter files. Sources: martinloop:readme.
- retention — Retention [partial]: Hosted Pro lists 30-day log retention; local retention is the user's. Sources: martinloop:pricing.
- integrity — Integrity [partial]: HMAC per-runs-root key detects tampering; integrity is evidence, not a universal guarantee. Sources: martinloop:readme.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:readme, martinloop:cli-ref, martinloop:mcp-tools, martinloop:rel090, martinloop:leash, martinloop:srch-team, martinloop:pricing

### TEAM-05: Review experience and guidance

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Review is terminal receipts: dossier, proof card, `martin_eval` grading and PR summaries/review helpers. No diff viewer or comment workflow in the product.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+), @martinloop/mcp server (stdio and HTTP transports); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- change_understanding: Dossier lists changed files, verifier steps and cost; no diff UI.
- evidence_navigation: `runs get/attempt/verify` navigate attempts and evidence.
- comments: No review comments.
- request_changes: No request-changes flow; `martin_review_pr` produces findings text.
- effort: No reviewer effort measurements.
- keyboard: Terminal CLI; no documented keyboard shortcuts.
- narrow_screen: Not documented.
- context_switching: `martin_pr_summary` and review helpers carry run evidence context.

- pending_reviews — Pending reviews [supported]: `triage` and `martin_triage_runs` rank runs by urgency and missing evidence. Sources: martinloop:mcp-tools, martinloop:readme.
- change_understanding — Change understanding [partial]: Dossier lists changed files, verifier steps and cost; no diff UI. Sources: martinloop:readme.
- evidence_navigation — Evidence navigation [partial]: `runs get/attempt/verify` navigate attempts and evidence. Sources: martinloop:cli-ref.
- comments — Comments [unsupported]: No review comments. Sources: martinloop:srch-team.
- request_changes — Request changes [unsupported]: No request-changes flow; `martin_review_pr` produces findings text. Sources: martinloop:prtools.
- shipping_guidance — Shipping guidance [supported]: Verified handoff states the next safe action. Sources: martinloop:readme.
- effort — Effort [unknown]: No reviewer effort measurements. Sources: martinloop:readme.
- missed_evidence — Missed evidence [supported]: Receipts flag EVIDENCE_BOUNDARY when rollback, budget or verifier evidence is missing. Sources: martinloop:readme.
- keyboard — Keyboard [unknown]: Terminal CLI; no documented keyboard shortcuts. Sources: martinloop:readme.
- narrow_screen — Narrow screen [unknown]: Not documented. Sources: martinloop:readme.
- context_switching — Context switching [partial]: `martin_pr_summary` and review helpers carry run evidence context. Sources: martinloop:mcp-tools.
- roles — Roles [partial]: Handoff names child/global verifier/parent authority separately in swarms. Sources: martinloop:hosted-handoff.
- retained_context — Retained context [supported]: Receipts retain verifier and budget context after the session. Sources: martinloop:readme.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:readme, martinloop:mcp-tools, martinloop:prtools, martinloop:cli-ref, martinloop:srch-team, martinloop:hosted-handoff

### TEAM-06: Colleague takeover and continuity

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Local single-user records; a second person can inspect a shared redacted receipt bundle but cannot take over a run, and hosted team workspaces are Growth waitlist.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+), Hosted dashboard / Atlas (closed, Pro plan); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- ongoing_takeover: No takeover; runs execute under one local user.
- same_task: `resume <loopId>` on the same machine reuses the same loop ID.
- session_history: Records persist locally; share bundle exports a redacted summary.
- decisions: Recorded in dossier.
- artifacts: Receipts and share bundle files.
- permissions: No shared permissions.
- runtime_context: Runtime context (checkout, agent session) stays on the local host.
- owner_disconnect: Local runs need the owner machine.

- ongoing_takeover — Ongoing takeover [unsupported]: No takeover; runs execute under one local user. Sources: martinloop:readme, martinloop:mcp-tools.
- same_task — Same task [partial]: `resume <loopId>` on the same machine reuses the same loop ID. Sources: martinloop:cli-ref.
- session_history — Session history [partial]: Records persist locally; share bundle exports a redacted summary. Sources: martinloop:readme.
- decisions — Decisions [partial]: Recorded in dossier. Sources: martinloop:readme.
- artifacts — Artifacts [partial]: Receipts and share bundle files. Sources: martinloop:readme.
- permissions — Permissions [unsupported]: No shared permissions. Sources: martinloop:srch-team.
- runtime_context — Runtime context [unsupported]: Runtime context (checkout, agent session) stays on the local host. Sources: martinloop:readme.
- owner_disconnect — Owner disconnect [unsupported]: Local runs need the owner machine. Sources: martinloop:readme.
- ownership — Ownership [unsupported]: No ownership transfer. Sources: martinloop:srch-team.
- attribution — Attribution [partial]: Local OS user principal only. Sources: martinloop:mcp-tools.
- context_loss — Context loss [partial]: A different person lacks the local checkout and agent session. Sources: martinloop:readme.
- harness_transfer — Harness transfer [partial]: Fallback chain moves a run to another configured engine under the same record. Sources: martinloop:changelog.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:readme, martinloop:pricing, martinloop:mcp-tools, martinloop:cli-ref, martinloop:srch-team, martinloop:changelog

## Verification and delivery

### SHIP-01: Retained artifacts and outputs

Verification status: Sourced (sourced)
Support: partial

Retained artifacts are run records, receipts, verifier results, rollback evidence, diffs in the run store and proof-card SVG/Markdown/JSON share bundles; no screenshots, recordings or gallery.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- diffs: Repo-backed runs can persist diffs and rollback artifacts when a persistence store is configured.
- documents: JSON and Markdown receipts under share/.
- screenshots: No screenshot capture.
- recordings: No recordings.
- previews: No previews.
- reviewer_outputs: `martin_review_pr` returns findings text; not stored as a retained reviewer artifact.
- gallery: No artifact gallery.
- revision_links: Receipts bind verifier evidence to the active run and workspace; commit linkage is in rollback boundaries.

- diffs — Diffs [partial]: Repo-backed runs can persist diffs and rollback artifacts when a persistence store is configured. Sources: martinloop:rollback.
- documents — Documents [partial]: JSON and Markdown receipts under share/. Sources: martinloop:readme.
- screenshots — Screenshots [unsupported]: No screenshot capture. Sources: martinloop:srch-clients, martinloop:rev-srch-media.
- recordings — Recordings [unsupported]: No recordings. Sources: martinloop:srch-clients, martinloop:rev-srch-media.
- previews — Previews [unsupported]: No previews. Sources: martinloop:srch-clients, martinloop:rev-srch-media.
- test_outputs — Test outputs [supported]: Verifier steps and output evidence are stored. Sources: martinloop:readme.
- reviewer_outputs — Reviewer outputs [partial]: `martin_review_pr` returns findings text; not stored as a retained reviewer artifact. Sources: martinloop:prtools.
- gallery — Gallery [unsupported]: No artifact gallery. Sources: martinloop:srch-clients, martinloop:rev-srch-media.
- task_links — Task links [supported]: Artifacts live under the run/loop. Sources: martinloop:readme.
- revision_links — Revision links [partial]: Receipts bind verifier evidence to the active run and workspace; commit linkage is in rollback boundaries. Sources: martinloop:readme, martinloop:rollback.
- persistence — Persistence [supported]: Persist under the runs root after sessions end. Sources: martinloop:readme.
- child_findings — Child findings [partial]: Swarm evidence keeps child, verifier and parent receipts. Sources: martinloop:hosted-handoff.
- prompt_leakage — Prompt leakage [supported]: Share bundles are redacted and the telemetry allowlist excludes prompts, task text and paths. Sources: martinloop:readme.

Checked 2026-11-16 (6 days old on 2026-11-22); not separately cross-checked.

Sources: martinloop:readme, martinloop:cli-ref, martinloop:rollback, martinloop:srch-clients, martinloop:prtools, martinloop:hosted-handoff, martinloop:rev-srch-media

### SHIP-02: Verification and spec gates

Verification status: Sourced (sourced)
Support: partial

Completion requires fresh verifier evidence from a user-configured command (tests, lint, types or anything); failed verification blocks VERIFIED and PR creation; no spec gates.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+), @martinloop/mcp server (stdio and HTTP transports); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- types: Any command; not built in.
- ci: GitHub Actions budget-gate example; no CI result ingestion.
- independent_review: Swarm global verifier is separate from child agents; `martin_review_pr` for PRs.
- spec_checks: No spec conformance checks.
- requirement_conflicts: No product requirement store.

- tests — Tests [supported]: `--verify "<cmd>"` after each attempt; verifier rules in config. Sources: martinloop:cli-ref, martinloop:config.
- lint — Lint [supported]: Any verifier command (config example lists pnpm lint); no built-in linter. Sources: martinloop:config.
- types — Types [partial]: Any command; not built in. Sources: martinloop:cli-ref.
- ci — Ci [partial]: GitHub Actions budget-gate example; no CI result ingestion. Sources: martinloop:examples, martinloop:srch-sched.
- independent_review — Independent review [partial]: Swarm global verifier is separate from child agents; `martin_review_pr` for PRs. Sources: martinloop:hosted-handoff, martinloop:prtools.
- spec_checks — Spec checks [unsupported]: No spec conformance checks. Sources: martinloop:readme, martinloop:rev-srch-spec.
- required_checks — Required checks [supported]: A run cannot be VERIFIED without verifier evidence; test-removal needs approval. Sources: martinloop:readme, martinloop:secpol.
- failure_blocking — Failure blocking [supported]: Failed verification yields no VERIFIED and PR creation refuses non-VERIFIED runs. Sources: martinloop:prtools, martinloop:readme.
- requirement_conflicts — Requirement conflicts [unsupported]: No product requirement store. Sources: martinloop:readme, martinloop:rev-srch-spec.
- configured_policy — Configured policy [supported]: martin.config.yaml verifierRules and policyProfile. Sources: martinloop:config.
- enforced_policy — Enforced policy [supported]: Enforced in run code and policy leash. Sources: martinloop:secpol, martinloop:leash.

Checked 2026-11-16 (6 days old on 2026-11-22); not separately cross-checked.

Sources: martinloop:readme, martinloop:cli-ref, martinloop:prtools, martinloop:secpol, martinloop:config, martinloop:examples, martinloop:srch-sched, martinloop:hosted-handoff, martinloop:leash, martinloop:rev-srch-spec

### SHIP-03: Git and pull request workflow

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Git is used for observation and rollback (reset --soft, snapshots) and swarm worktrees; PR creation only through `martin_create_pr` using gh. No commits, branches, review comments, CI response or rebase.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+), @martinloop/mcp server (stdio and HTTP transports); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- branches: No branch creation; reads current branch for PR.
- commits: Rewinds agent commits via git reset --soft when blocked; does not author commits itself.
- comments: No PR comment posting found.
- ci_response: No CI response handling.
- rebase: No rebase code.
- conflicts: Swarm integration workspace integrates child results; conflict handling beyond failure is not documented.
- product_state: Run records keep verifier and rollback state; PR URL is returned, not stored.

- branches — Branches [unsupported]: No branch creation; reads current branch for PR. Sources: martinloop:prtools, martinloop:srch-merge.
- worktrees — Worktrees [supported]: Swarm child, integration and verifier workspaces. Sources: martinloop:workspaces.
- commits — Commits [partial]: Rewinds agent commits via git reset --soft when blocked; does not author commits itself. Sources: martinloop:rollback.
- pull_requests — Pull requests [supported]: `martin_create_pr` runs `gh pr create` for a VERIFIED run (execute=true). Sources: martinloop:prtools.
- comments — Comments [unsupported]: No PR comment posting found. Sources: martinloop:srch-merge.
- ci_response — Ci response [unsupported]: No CI response handling. Sources: martinloop:srch-merge.
- rebase — Rebase [unsupported]: No rebase code. Sources: martinloop:srch-merge.
- conflicts — Conflicts [partial]: Swarm integration workspace integrates child results; conflict handling beyond failure is not documented. Sources: martinloop:workspaces.
- product_state — Product state [partial]: Run records keep verifier and rollback state; PR URL is returned, not stored. Sources: martinloop:prtools.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:rollback, martinloop:workspaces, martinloop:prtools, martinloop:srch-merge

### SHIP-04: Merge and deployment authority

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

MartinLoop can open a PR only from a verified run; it does not merge, release or deploy.

Availability: generally-available
Delivery: native
Scope: surfaces=@martinloop/mcp server (stdio and HTTP transports); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- approve: Verified handoff gates PR opening; no reviewer approval flow.
- merge: No merge code.
- release: No release tooling.
- deploy: No deployment tooling.
- exact_revision: Receipts bind verifier evidence to the active run and workspace.
- authority_enforcement: `martin_create_pr` refuses non-VERIFIED or integrity-failed runs.

- approve — Approve [partial]: Verified handoff gates PR opening; no reviewer approval flow. Sources: martinloop:prtools.
- merge — Merge [unsupported]: No merge code. Sources: martinloop:srch-merge.
- release — Release [unsupported]: No release tooling. Sources: martinloop:srch-merge.
- deploy — Deploy [unsupported]: No deployment tooling. Sources: martinloop:srch-merge.
- exact_revision — Exact revision [partial]: Receipts bind verifier evidence to the active run and workspace. Sources: martinloop:readme.
- authority_enforcement — Authority enforcement [partial]: `martin_create_pr` refuses non-VERIFIED or integrity-failed runs. Sources: martinloop:prtools.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:prtools, martinloop:srch-merge, martinloop:readme

### SHIP-05: Completion evidence and rollback

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Receipts link verifier evidence, changed files and rollback outcomes to the run; no commit/PR/build/deployment link store; rollback restores the workspace.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- commit_links: Rollback boundaries record the head commit at attempt start.
- pr_links: PR creation returns a URL; not persisted in the run record.
- build_links: No build links.
- deployment_links: No deployment links.
- delivery_status: VERIFIED/STOPPED/NEEDS REVIEW describe run outcome, not delivery.
- revert: Restores workspace to pre-attempt state; no revert of merged PRs.
- stale_approval_invalidation: Verifier evidence must be fresh and bound to the active run and workspace.

- commit_links — Commit links [partial]: Rollback boundaries record the head commit at attempt start. Sources: martinloop:rollback.
- pr_links — Pr links [partial]: PR creation returns a URL; not persisted in the run record. Sources: martinloop:prtools.
- build_links — Build links [unsupported]: No build links. Sources: martinloop:srch-merge.
- deployment_links — Deployment links [unsupported]: No deployment links. Sources: martinloop:srch-merge.
- delivery_status — Delivery status [partial]: VERIFIED/STOPPED/NEEDS REVIEW describe run outcome, not delivery. Sources: martinloop:readme.
- retained_evidence — Retained evidence [supported]: Receipts and dossiers retain evidence. Sources: martinloop:readme.
- rollback — Rollback [supported]: Rollback boundary restore after failed/blocked attempts with explicit failed/unavailable states. Sources: martinloop:rollback, martinloop:secpol.
- revert — Revert [partial]: Restores workspace to pre-attempt state; no revert of merged PRs. Sources: martinloop:rollback.
- stale_approval_invalidation — Stale approval invalidation [partial]: Verifier evidence must be fresh and bound to the active run and workspace. Sources: martinloop:readme.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:readme, martinloop:rollback, martinloop:prtools, martinloop:srch-merge, martinloop:secpol

### SHIP-06: Batch review and high-volume shipping

Verification status: Sourced (sourced)
Support: partial

Terminal triage ranking of saved runs and per-run dossiers; no batch review queue, grouping or merge queue.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- review_queue: `triage` ranks saved runs by urgency and missing evidence.
- grouping: No grouping.
- risk_size_filters: `martin_eval` grades risk and reviewability per run; no filter UI.
- findings: Findings text from martin_review_pr.
- artifacts: Share bundles per run.
- request_changes: No request-changes flow.
- exact_revision_approval: No approval step.
- dependency_queues: No dependency-aware merge queue.

- review_queue — Review queue [partial]: `triage` ranks saved runs by urgency and missing evidence. Sources: martinloop:cli-ref, martinloop:mcp-tools.
- summaries — Summaries [supported]: Dossiers and martin_pr_summary summarize a run. Sources: martinloop:mcp-tools.
- grouping — Grouping [unsupported]: No grouping. Sources: martinloop:cli-ref, martinloop:rev-srch-group.
- risk_size_filters — Risk size filters [partial]: `martin_eval` grades risk and reviewability per run; no filter UI. Sources: martinloop:mcp-tools.
- findings — Findings [partial]: Findings text from martin_review_pr. Sources: martinloop:prtools.
- artifacts — Artifacts [partial]: Share bundles per run. Sources: martinloop:readme.
- request_changes — Request changes [unsupported]: No request-changes flow. Sources: martinloop:prtools.
- exact_revision_approval — Exact revision approval [unsupported]: No approval step. Sources: martinloop:prtools.
- dependency_queues — Dependency queues [unsupported]: No dependency-aware merge queue. Sources: martinloop:srch-merge.
- next_actions — Next actions [supported]: Verified handoff names the next safe action. Sources: martinloop:readme.
- stale_approval_invalidation — Stale approval invalidation [unsupported]: No approvals to invalidate. Sources: martinloop:prtools.
- check_deduplication — Check deduplication [unsupported]: No check deduplication. Sources: martinloop:scheduler.

Checked 2026-11-16 (6 days old on 2026-11-22); not separately cross-checked.

Sources: martinloop:cli-ref, martinloop:mcp-tools, martinloop:prtools, martinloop:readme, martinloop:srch-merge, martinloop:scheduler, martinloop:rev-srch-group

## Ownership, privacy and economics

### OWN-01: Source and license

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Core (CLI, MCP, SDK, adapters, plugin) is Apache-2.0 at v0.9.0; hosted product surfaces (dashboard, Atlas, control plane) are proprietary commercial and not in the repository.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+), Hosted dashboard / Atlas (closed, Pro plan); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- hosted_parity: Hosted dashboard features are not in the repository; parity cannot be assumed.

- client_license — Client license [supported]: Apache-2.0 (LICENSE; package.json license). Sources: martinloop:repoapi, martinloop:readme.
- server_license — Server license [supported]: MCP server is part of the Apache-2.0 repo. Sources: martinloop:readme.
- hosted_extensions — Hosted extensions [supported]: llms.txt: hosted product surfaces are proprietary commercial. Sources: martinloop:llms.
- source_visibility — Source visibility [supported]: Public GitHub repository; REUSE.toml present. Sources: martinloop:repoapi.
- source_revision — Source revision [supported]: Pinned to tag v0.9.0, commit df38850. Sources: martinloop:repoapi.
- hosted_parity — Hosted parity [unsupported]: Hosted dashboard features are not in the repository; parity cannot be assumed. Sources: martinloop:llms, martinloop:hosted-handoff.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:repoapi, martinloop:readme, martinloop:llms, martinloop:hosted-handoff

### OWN-02: Hosting and exit

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Core is fully local with JSONL run store and no vendor account; hosted sync is optional. Enterprise VPC/on-prem is contact-sales. Exports are JSON/Markdown/SVG receipts.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+), Hosted dashboard / Atlas (closed, Pro plan); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- self_hosting: Core runs locally; the hosted dashboard is closed; Enterprise VPC/on-prem is contact-sales.
- import: No import command for run records found.
- cancellation: Hosted Pro cancellation terms are in closed terms/billing pages not read.
- history_portability: Run records are portable files; hosted history is not exportable per documentation read.

- self_hosting — Self hosting [partial]: Core runs locally; the hosted dashboard is closed; Enterprise VPC/on-prem is contact-sales. Sources: martinloop:readme, martinloop:pricing.
- vendor_dependencies — Vendor dependencies [supported]: Core needs none beyond the chosen agent CLI/provider; telemetry and sync are optional. Sources: martinloop:readme.
- export — Export [supported]: `share`, `dossier --json`, JSONL records. Sources: martinloop:cli-ref, martinloop:readme.
- import — Import [unsupported]: No import command for run records found. Sources: martinloop:cli-ref.
- formats — Formats [supported]: JSONL, JSON, Markdown, SVG. Sources: martinloop:readme.
- cancellation — Cancellation [unknown]: Hosted Pro cancellation terms are in closed terms/billing pages not read. Sources: martinloop:pricing.
- account_independence — Account independence [supported]: Core works without any account and with telemetry disabled. Sources: martinloop:readme.
- history_portability — History portability [partial]: Run records are portable files; hosted history is not exportable per documentation read. Sources: martinloop:readme, martinloop:pricing.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:readme, martinloop:cli-ref, martinloop:pricing, martinloop:hosted-handoff

### OWN-03: Data handling and telemetry

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Local-first: code and prompts stay local; hosted sync uploads preserved evidence only when configured. Telemetry is anonymous, allowlisted, on by default after a one-time disclosure and controllable. Providers see code per their own policy.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+), Hosted dashboard / Atlas (closed, Pro plan); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- logs: Run logs stay local unless `sync flush` is configured.
- artifacts: Share bundle is redacted and user-distributed.
- retention: Hosted Pro: 30-day log retention; local files are the user's.
- deletion: Hosted deletion process is not documented in read sources.
- training: Hosted training-use policy is in closed privacy terms not read; provider policy applies to code sent to agents.
- residency: No residency statement for the hosted service; telemetry endpoint is a Supabase function.
- subprocessors: No subprocessor list read; telemetry names a Supabase endpoint.
- consent_prompt: One-time non-blocking disclosure, not a consent prompt; blocked until shown.

- code_data — Code data [supported]: Source code, prompts and file paths are never sent in telemetry; agent providers receive code per their policies. Sources: martinloop:readme, martinloop:telemetry.
- prompts — Prompts [supported]: Prompts and task text excluded from telemetry; hosted sync sends allowlisted evidence. Sources: martinloop:readme, martinloop:hosted-handoff.
- credentials — Credentials [supported]: Secrets and env vars excluded from telemetry; tokens stay in environment; secret-like task text refused. Sources: martinloop:readme, martinloop:secpol.
- logs — Logs [partial]: Run logs stay local unless `sync flush` is configured. Sources: martinloop:cli-ref.
- artifacts — Artifacts [partial]: Share bundle is redacted and user-distributed. Sources: martinloop:readme.
- retention — Retention [partial]: Hosted Pro: 30-day log retention; local files are the user's. Sources: martinloop:pricing.
- deletion — Deletion [unknown]: Hosted deletion process is not documented in read sources. Sources: martinloop:pricing.
- training — Training [unknown]: Hosted training-use policy is in closed privacy terms not read; provider policy applies to code sent to agents. Sources: martinloop:llms.
- residency — Residency [unknown]: No residency statement for the hosted service; telemetry endpoint is a Supabase function. Sources: martinloop:readme.
- subprocessors — Subprocessors [unknown]: No subprocessor list read; telemetry names a Supabase endpoint. Sources: martinloop:readme.
- telemetry_disclosure — Telemetry disclosure [supported]: README lists what is and is not sent and the endpoint. Sources: martinloop:readme.
- telemetry_defaults — Telemetry defaults [supported]: Enabled by default after a one-time disclosure on interactive installs. Sources: martinloop:readme, martinloop:telemetry.
- telemetry_controls — Telemetry controls [supported]: `martin telemetry off`, DO_NOT_TRACK, MARTIN_TELEMETRY_DISABLED, CI=1, `telemetry explain`. Sources: martinloop:readme, martinloop:telemetry.
- consent_prompt — Consent prompt [partial]: One-time non-blocking disclosure, not a consent prompt; blocked until shown. Sources: martinloop:telemetry.
- provider_separation — Provider separation [supported]: Provider/model policies are separate from MartinLoop telemetry. Sources: martinloop:models, martinloop:readme.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:readme, martinloop:telemetry, martinloop:cli-ref, martinloop:hosted-handoff, martinloop:secpol, martinloop:pricing, martinloop:llms, martinloop:models

### OWN-04: Price and included usage

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Core is free (Apache-2.0). Hosted Pro is $49/month billed monthly (dashboard, run receipts, cost history, 30-day log retention, personal projects); Growth is waitlist-only and Enterprise is contact-sales. Model costs are separate.

Availability: generally-available
Delivery: native
Scope: surfaces=Hosted dashboard / Atlas (closed, Pro plan), martin-loop CLI (npm package martin-loop, Node 20+); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core; hosted Pro $49/month; os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- seats: Pro targets solo users with personal projects; per-seat pricing for Growth is not stated.
- seat_minimum: Not stated for Growth/Enterprise.
- workspace_fees: No workspace fee stated.
- included_usage: Pro includes hosted dashboard, receipts, cost history and 30-day retention; no usage quotas are stated.
- credits: No model credits.
- trial: No trial stated on the pricing page text read.
- overages: No overage terms stated.
- taxes: Tax treatment not stated.

- currency — Currency [supported]: USD on the pricing page. Sources: martinloop:pricing.
- interval — Interval [supported]: Pro is billed monthly. Sources: martinloop:pricing.
- seats — Seats [partial]: Pro targets solo users with personal projects; per-seat pricing for Growth is not stated. Sources: martinloop:pricing.
- seat_minimum — Seat minimum [unknown]: Not stated for Growth/Enterprise. Sources: martinloop:pricing.
- workspace_fees — Workspace fees [unknown]: No workspace fee stated. Sources: martinloop:pricing.
- included_usage — Included usage [partial]: Pro includes hosted dashboard, receipts, cost history and 30-day retention; no usage quotas are stated. Sources: martinloop:pricing.
- credits — Credits [unsupported]: No model credits. Sources: martinloop:pricing.
- trial — Trial [unknown]: No trial stated on the pricing page text read. Sources: martinloop:pricing.
- overages — Overages [unknown]: No overage terms stated. Sources: martinloop:pricing.
- taxes — Taxes [unknown]: Tax treatment not stated. Sources: martinloop:pricing.
- model_cost — Model cost [supported]: Model/provider cost is separate; MartinLoop reports it. Sources: martinloop:readme, martinloop:models.
- compute_cost — Compute cost [supported]: No compute fees; local execution. Sources: martinloop:readme.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:pricing, martinloop:llms, martinloop:readme, martinloop:models

### OWN-05: Scenario cost and controls

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Core costs $0 for any number of developers; hosted Pro would be $49/month per user if each developer subscribed (personal projects only). Agent/model costs depend on the user's provider and are not modeled here.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+), Hosted dashboard / Atlas (closed, Pro plan); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- one_person: Core $0; with Pro $49/month; model cost excluded.
- two_people: Core $0 each; Pro is solo/personal projects and shared team dashboards are waitlist, so two collaborators cannot share one Pro workspace.
- five_people: Core $0 each; five Pro subscriptions would be 5 × $49 = $245/month with no shared dashboard.
- task_assumptions: No fixed workload is published; the catalog invents none.
- tokens: Token spend is provider-specific.
- variable_cost: Model/provider usage varies.
- included_cost: Pro includes hosted dashboard features only.
- unknown_cost: Growth/Enterprise prices are not published.

- one_person — One person [partial]: Core $0; with Pro $49/month; model cost excluded. Sources: martinloop:pricing.
- two_people — Two people [partial]: Core $0 each; Pro is solo/personal projects and shared team dashboards are waitlist, so two collaborators cannot share one Pro workspace. Sources: martinloop:pricing.
- five_people — Five people [partial]: Core $0 each; five Pro subscriptions would be 5 × $49 = $245/month with no shared dashboard. Sources: martinloop:pricing.
- task_assumptions — Task assumptions [unknown]: No fixed workload is published; the catalog invents none. Sources: martinloop:readme.
- concurrency — Concurrency [supported]: Swarm concurrency is set by the user; no pricing impact. Sources: martinloop:rel090.
- tokens — Tokens [unknown]: Token spend is provider-specific. Sources: martinloop:models.
- compute_hours — Compute hours [supported]: No compute-hours charge. Sources: martinloop:readme.
- fixed_cost — Fixed cost [supported]: Pro $49/month fixed; Core $0. Sources: martinloop:pricing.
- variable_cost — Variable cost [unknown]: Model/provider usage varies. Sources: martinloop:models.
- included_cost — Included cost [partial]: Pro includes hosted dashboard features only. Sources: martinloop:pricing.
- unknown_cost — Unknown cost [unknown]: Growth/Enterprise prices are not published. Sources: martinloop:pricing.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:pricing, martinloop:readme, martinloop:rel090, martinloop:models

### OWN-06: Availability and support

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Pre-1.0 public releases every few days (v0.9.0 on 2026-11-03); support via GitHub issues and support@martinloop.com; no response commitments or status page found; hosted Growth is waitlist.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+), Hosted dashboard / Atlas (closed, Pro plan); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- release_stage: Version 0.9.0 pre-1.0; hosted Pro generally available per pricing page, Growth waitlist.
- access_restrictions: Growth waitlist and Enterprise contact-sales restrict access.
- response_commitments: Enterprise lists an SLA but no terms are public.
- status_history: No status page found.
- deprecation: No deprecation policy found.

- release_stage — Release stage [partial]: Version 0.9.0 pre-1.0; hosted Pro generally available per pricing page, Growth waitlist. Sources: martinloop:repoapi, martinloop:pricing.
- access_restrictions — Access restrictions [partial]: Growth waitlist and Enterprise contact-sales restrict access. Sources: martinloop:pricing.
- support — Support [supported]: GitHub issues and support@martinloop.com; Enterprise lists dedicated support. Sources: martinloop:readme, martinloop:pricing.
- response_commitments — Response commitments [unknown]: Enterprise lists an SLA but no terms are public. Sources: martinloop:pricing.
- release_activity — Release activity [supported]: Releases v0.8.9 (2026-10-22), v0.8.10 (2026-10-26) and v0.9.0 (2026-11-03). Sources: martinloop:changelog, martinloop:repoapi.
- status_history — Status history [unknown]: No status page found. Sources: martinloop:readme.
- deprecation — Deprecation [unknown]: No deprecation policy found. Sources: martinloop:readme.
- export_options — Export options [supported]: Receipts and JSONL exports. Sources: martinloop:readme.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:repoapi, martinloop:changelog, martinloop:readme, martinloop:pricing, martinloop:llms

### OWN-07: Enterprise assurance

Verification status: No sourced public finding (not-published)
Support: not-published

No publishable sourced finding is available for this criterion. This does not mean the product lacks the capability.

### OWN-08: Useful free tier and evaluation access

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Core is free and open source with no account; hosted Pro is paid ($49/month) with no stated trial; Growth is waitlist-only. Two-person collaboration, mobile and shared review are not free scenarios because they do not exist in Core.

Availability: generally-available
Delivery: native
Scope: surfaces=martin-loop CLI (npm package martin-loop, Node 20+), Hosted dashboard / Atlas (closed, Pro plan); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- trial: No Pro trial stated.
- credits_trial: No credits trial.
- paid_only: Hosted dashboard is paid-only (Pro $49/month); Core is not.
- early_access: Growth is waitlist only; Enterprise via sales.
- free_scenarios: Free: local governed runs, BYO subscriptions, receipts and swarm demo; no free collaboration, remote compute, mobile or issue management.
- card_required: Not stated for Core; Pro is checkout-based.

- ongoing_free — Ongoing free [supported]: Open Source Core is free "forever" per the pricing page. Sources: martinloop:pricing.
- trial — Trial [unknown]: No Pro trial stated. Sources: martinloop:pricing.
- credits_trial — Credits trial [unsupported]: No credits trial. Sources: martinloop:pricing.
- self_host_cost — Self host cost [supported]: Local execution; user pays providers only. Sources: martinloop:readme.
- paid_only — Paid only [partial]: Hosted dashboard is paid-only (Pro $49/month); Core is not. Sources: martinloop:pricing.
- early_access — Early access [partial]: Growth is waitlist only; Enterprise via sales. Sources: martinloop:pricing.
- free_scenarios — Free scenarios [partial]: Free: local governed runs, BYO subscriptions, receipts and swarm demo; no free collaboration, remote compute, mobile or issue management. Sources: martinloop:readme, martinloop:pricing.
- card_required — Card required [unknown]: Not stated for Core; Pro is checkout-based. Sources: martinloop:pricing.
- usage_caps — Usage caps [supported]: No caps on Core. Sources: martinloop:readme.
- paid_evaluation — Paid evaluation [supported]: Meaningful evaluation of the CLI needs no payment. Sources: martinloop:readme.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:pricing, martinloop:readme

### OWN-09: Pricing transparency

Verification status: Agent cross-checked (agent-cross-checked)
Support: partial

Free Core and Pro $49/month are public; Growth is unpriced (waitlist) and Enterprise is quote-only. Overage, trial and seat terms are not stated.

Availability: generally-available
Delivery: native
Scope: surfaces=Hosted dashboard / Atlas (closed, Pro plan); version=v0.9.0 (commit df38850, released 2026-11-03); plan=Open-source Core (Apache-2.0); os=not specified; execution_modes=not specified; region=not specified; harness=not specified.

Limitations:
- units: Pro is per month for solo use; Growth/Enterprise units unstated.
- included_limits: Pro lists features and 30-day retention only.
- overages: No overage terms.
- minimum_commitment: Not stated.
- cancellation: Cancellation terms are not shown on the pricing page text.
- sales_only_unknowns: Enterprise is contact-sales; Growth waitlist.
- transparency_classification: Partially specified: Core and Pro priced; Growth and Enterprise unpriced.

- units — Units [partial]: Pro is per month for solo use; Growth/Enterprise units unstated. Sources: martinloop:pricing.
- included_limits — Included limits [partial]: Pro lists features and 30-day retention only. Sources: martinloop:pricing.
- overages — Overages [unknown]: No overage terms. Sources: martinloop:pricing.
- compute_markup — Compute markup [supported]: No compute is sold. Sources: martinloop:readme.
- model_markup — Model markup [supported]: Model providers bill separately with no markup by MartinLoop. Sources: martinloop:models.
- minimum_commitment — Minimum commitment [unknown]: Not stated. Sources: martinloop:pricing.
- cancellation — Cancellation [unknown]: Cancellation terms are not shown on the pricing page text. Sources: martinloop:pricing.
- sales_only_unknowns — Sales only unknowns [partial]: Enterprise is contact-sales; Growth waitlist. Sources: martinloop:pricing.
- transparency_classification — Transparency classification [partial]: Partially specified: Core and Pro priced; Growth and Enterprise unpriced. Sources: martinloop:pricing.

Checked 2026-11-16 (6 days old on 2026-11-22); agent cross-checked 2026-11-17.

Sources: martinloop:pricing, martinloop:readme, martinloop:models

## Sources

### martinloop:readme

MartinLoop README. Governed-run CLI with budget caps, verifier gates, receipts, 13-class failure taxonomy, swarm mode demo, MCP server, SDK, local-first storage, optional hosted sync and telemetry with opt-out.

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/README.md
- Locator: Start Here, Model and Engine Support, CLI, MCP, SDK, Telemetry & Privacy
- Evidence kind: documented
- Source date: Not stated
- Retrieved: 2026-11-16

### martinloop:cli-ref

CLI reference. Command list (start, doctor, run, preflight, dossier, resume, share, runs, sync, mcp) and run options for budget, iterations, tokens, engine, allow/deny paths and verifier.

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/docs/reference/cli.md
- Locator: Commands, Run Options, Hosted sync
- Evidence kind: documented
- Source date: Not stated
- Retrieved: 2026-11-16

### martinloop:mcp-tools

MCP tool reference. Tools for plan, run, status, pause, cancel, continue, inspect, dossier, PR summary/create/review; run control only over local stdio with a live executor.

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/docs/reference/mcp-tools.md
- Locator: tool tables, run-control paragraph
- Evidence kind: documented
- Source date: Not stated
- Retrieved: 2026-11-16

### martinloop:leash

Safety leash. Pre-execution checks on commands, paths, secrets, spend, network and dependency/migration/config/test changes; approvals are explicit flags.

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/packages/core/src/leash.ts
- Locator: SafetySurface, SafetyViolationKind, execution profiles
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-16
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages/core/src/leash.ts

### martinloop:hosted-handoff

Swarm hosted v1 consumer contract. Defines the swarm sync envelope for a closed native control plane; states no ingestion or dashboard surface is claimed in this repository.

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/docs/reference/SWARM-HOSTED-V1-HANDOFF.md
- Locator: Status and boundary; Tenant and route ownership
- Evidence kind: documented
- Source date: Not stated
- Retrieved: 2026-11-16

### martinloop:llms

MartinLoop llms.txt. States supported agents Claude Code, Codex, Gemini CLI; Core is Apache-2.0 and hosted product surfaces are proprietary commercial; Atlas is a paid tenant-scoped surface.

- URL: https://martinloop.com/llms.txt
- Locator: Key facts, License, Core capabilities
- Evidence kind: documented
- Source date: Not stated
- Retrieved: 2026-11-16

### martinloop:pricing

MartinLoop pricing. Open Source Core free; Pro $49/month billed monthly (hosted dashboard, run receipts, cost history, 30-day log retention, personal projects); Growth waitlist only; Enterprise contact sales (VPC/on-prem, RBAC + SSO, audit-log retention).

- URL: https://martinloop.com/pricing
- Locator: plan cards rendered from the published site bundle (assets/index-CBl-S_PX.js)
- Evidence kind: documented
- Source date: Not stated
- Retrieved: 2026-11-16

### martinloop:release-ci

Platform release validation workflow. Hosted Windows, macOS and Linux validation matrix for release candidates.

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/.github/workflows/platform-release-validation.yml
- Locator: matrix os
- Evidence kind: documented
- Source date: Not stated
- Retrieved: 2026-11-16

### martinloop:plugin

Claude plugin manifest. Plugin metadata (name martinloop, 0.9.0, Apache-2.0) with a martinloop-govern skill.

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/plugins/martinloop/plugin.json
- Locator: plugin.json and skills/martinloop-govern
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-16
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: plugins/martinloop/plugin.json

### martinloop:srch-clients

Bounded source search. No mobile or desktop app code; only a CLI, MCP server and SDK ship (matches are a CLI bridge that avoids the bundled Electron Node when hosted inside Claude Code desktop or VS Code, and license text).

- URL: https://github.com/Keesan12/martin-loop/tree/df388508ac1451a7c3feb12b86ee424c60566208/packages
- Locator: rg -i "ios|android|electron|tauri|react-native" packages/*/src plugins scripts
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-16
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages

### martinloop:srch-notify

Bounded source search. No notification code in the pinned packages.

- URL: https://github.com/Keesan12/martin-loop/tree/df388508ac1451a7c3feb12b86ee424c60566208/packages
- Locator: rg -i "notify-send|osascript|desktop notification|node-notifier|toast" packages/*/src
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-16
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages

### martinloop:models

Coding agent and model support. Engines auto, codex, claude, gemini and openai (any OpenAI-compatible endpoint incl. Ollama/LM Studio); worker changes without changing the job contract.

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/docs/reference/model-support.md
- Locator: Engine contract, OpenAI-compatible models
- Evidence kind: documented
- Source date: Not stated
- Retrieved: 2026-11-16

### martinloop:changelog

Changelog. continue for interrupted reversible runs, provider fallback chain, durable execution leases, retained rollback authority across reconnects.

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/CHANGELOG.md
- Locator: 0.9.0, 0.8.10, 0.8.9 sections
- Evidence kind: documented
- Source date: Not stated
- Retrieved: 2026-11-16

### martinloop:examples

Examples. Example CI budget gate and a custom OpenAI-compatible adapter run record; manual integrations.

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/examples/README.md
- Locator: github-actions-budget-gate, opencode-adapter
- Evidence kind: documented
- Source date: Not stated
- Retrieved: 2026-11-16

### martinloop:secpol

Security policy model. Path, secret, test-change, dependency and verifier-command checks with rollback; explicitly not a sandbox for the agent process.

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/docs/security/policy.md
- Locator: Policy checks, file checks, Limits
- Evidence kind: documented
- Source date: Not stated
- Retrieved: 2026-11-16

### martinloop:rollback

Rollback boundaries. Records a pre-attempt boundary, restores files after blocked or failed attempts and rewinds committed changes; reports failed/unavailable rollback.

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/packages/core/src/rollback.ts
- Locator: git reset --soft, snapshots, ignored-file stamps
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-16
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages/core/src/rollback.ts

### martinloop:prtools

MCP PR tools. martin_create_pr runs `gh pr create` only when execute=true, the run is VERIFIED and receipt integrity is verified; otherwise previews.

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/packages/mcp/src/tools/pr-tools.ts
- Locator: martinCreatePrTool lines 60-120
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-16
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages/mcp/src/tools/pr-tools.ts

### martinloop:srch-isolation

Bounded source search. No container, VM or OS-sandbox launcher; docker appears only in destructive-command and config-file patterns.

- URL: https://github.com/Keesan12/martin-loop/tree/df388508ac1451a7c3feb12b86ee424c60566208/packages
- Locator: rg -i "docker|bubblewrap|bwrap|firejail|seatbelt|chroot|firecracker" packages/*/src
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-16
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages

### martinloop:rel090

MartinLoop 0.9.0 release notes. Governed Swarm needs explicit agent and concurrency limits; MCP pause/continue/cancel; crash-safe cleanup; not an OS sandbox.

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/docs/release/OSS-0.9.0-RELEASE-NOTES.md
- Locator: What changed, Safety boundary
- Evidence kind: documented
- Source date: Not stated
- Retrieved: 2026-11-16

### martinloop:mcp-server

MCP server. StdioServerTransport and an http transport option; run control gated on controlTransport.

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/packages/mcp/src/server.ts
- Locator: lines 2325-2360 transports
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-16
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages/mcp/src/server.ts

### martinloop:srch-sched

Bounded source search. No scheduler, cron or webhook receiver beyond swarm task scheduling and a CI example.

- URL: https://github.com/Keesan12/martin-loop/tree/df388508ac1451a7c3feb12b86ee424c60566208/packages
- Locator: rg -i "cron|setInterval|webhook|workflow_dispatch" packages/*/src
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-16
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages

### martinloop:srch-team

Bounded source search. No role, SSO or multi-user code in the CLI/core/MCP packages; hits are remote-experience invite notices and repo-analyzer wording.

- URL: https://github.com/Keesan12/martin-loop/tree/df388508ac1451a7c3feb12b86ee424c60566208/packages
- Locator: rg -i "rbac|sso|saml|oidc|invite|organization" packages/*/src
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-16
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages

### martinloop:rev-srch-mcpclient

Reviewer search: MCP client code. Reviewer bounded search at pinned commit: no hits.

- URL: https://github.com/Keesan12/martin-loop/tree/df388508ac1451a7c3feb12b86ee424c60566208
- Locator: grep -rIlE -i '@modelcontextprotocol/sdk/client|new Client\(' packages/cli/src packages/core/src packages/mcp/src packages/adapters/src packages/contracts/src (tests, fixtures and lockfiles excluded)
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-17
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages/cli/src

### martinloop:srch-tracker

Bounded source search. No tracker integration or issue entity in the pinned packages.

- URL: https://github.com/Keesan12/martin-loop/tree/df388508ac1451a7c3feb12b86ee424c60566208/packages
- Locator: rg -i "linear\.app|jira|issue tracker|backlog" packages/*/src
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-16
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages

### martinloop:rev-srch-edit

Reviewer search: run edit, archive or import. Reviewer bounded search at pinned commit: no hits.

- URL: https://github.com/Keesan12/martin-loop/tree/df388508ac1451a7c3feb12b86ee424c60566208
- Locator: grep -rIlE -i 'archiveRun|runs archive|updateRun|editRun|runs edit|runs delete|import-run|runs import' packages/cli/src packages/core/src packages/mcp/src packages/adapters/src packages/contracts/src plugins scripts (tests, fixtures and lockfiles excluded)
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-17
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages/cli/src

### martinloop:scheduler

Swarm scheduler. Tasks with dependsOn become ready only when dependencies are verified; effective concurrency is min(requested, parent maxConcurrency).

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/packages/core/src/swarm/scheduler.ts
- Locator: lines 360-480 dependsOn, maxConcurrency
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-16
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages/core/src/swarm/scheduler.ts

### martinloop:swarm-cli

Swarm CLI parser. swarm plan/run read a plan file; status (with --watch), inspect, cancel, dossier, verify and share operate on persisted swarm runs.

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/packages/cli/src/swarm-command-private.ts
- Locator: lines 539-700 plan/run/status/inspect/cancel
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-16
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages/cli/src/swarm-command-private.ts

### martinloop:config

Example config. policyProfile, budget (maxUsd, softLimitUsd, maxIterations, maxTokens), governance (destructiveActionPolicy approval, telemetryDestination local-only, verifierRules).

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/martin.config.example.yaml
- Locator: whole file
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-16
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: martin.config.example.yaml

### martinloop:memory

Memory store. Append-only JSONL memory of consent, preference, pattern, feedback and budget entries stored under the runs root.

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/packages/core/src/persistence/memory-store.ts
- Locator: header comment
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-16
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages/core/src/persistence/memory-store.ts

### martinloop:rev-srch-index

Reviewer search: code index or retrieval. Reviewer bounded search at pinned commit: no hits.

- URL: https://github.com/Keesan12/martin-loop/tree/df388508ac1451a7c3feb12b86ee424c60566208
- Locator: grep -rIlE -i 'embedding|vector store|semantic search|code index|knowledge base|tree-sitter' packages/cli/src packages/core/src packages/mcp/src packages/adapters/src packages/contracts/src plugins scripts (tests, fixtures and lockfiles excluded)
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-17
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages/cli/src

### martinloop:rev-srch-image

Reviewer search: image input. Reviewer bounded search at pinned commit: 2 file hit(s), reviewed: packages/contracts/src/index.ts, plugins/martinloop/.codex-plugin/plugin.json. The two hits are an artifact-kind union member ('screenshot', contracts/src/index.ts line 274) and a plugin listing field (.codex-plugin/plugin.json line 33); neither is image input handling or capture code.

- URL: https://github.com/Keesan12/martin-loop/tree/df388508ac1451a7c3feb12b86ee424c60566208
- Locator: grep -rIlE -i 'image_url|multimodal|screenshot|\bvision\b' packages/cli/src packages/core/src packages/mcp/src packages/adapters/src packages/contracts/src plugins scripts (tests, fixtures and lockfiles excluded)
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-17
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages/cli/src

### martinloop:rev-srch-spec

Reviewer search: living specification code. Reviewer bounded search at pinned commit: no hits.

- URL: https://github.com/Keesan12/martin-loop/tree/df388508ac1451a7c3feb12b86ee424c60566208
- Locator: grep -rIlE -i 'living spec|requirements? (store|registry|traceab)|spec(ification)? drift|product spec|\bPRD\b|requirement-to' packages/cli/src packages/core/src packages/mcp/src packages/adapters/src packages/contracts/src plugins scripts (tests, fixtures and lockfiles excluded)
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-17
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages/cli/src

### martinloop:rev-live-engine

Swarm live runtime: one parent-selected engine and model. A live swarm plan carries a single engine and model; the runtime throws if a child adapter reports a different provider or model ("every child must use the parent-selected provider/model").

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/packages/core/src/swarm/live-runtime.ts
- Locator: live-runtime.ts lines 725-741 (LIVE_ENGINE_PROVIDER_MISMATCH, LIVE_ENGINE_MODEL_MISMATCH); contracts/src/swarm.ts lines 417-431 (SwarmLiveEngineProfile, SwarmLivePlanDraft.engine)
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-17
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages/core/src/swarm/live-runtime.ts

### martinloop:rev-srch-steer

Reviewer search: live steering. Reviewer bounded search at pinned commit: no hits.

- URL: https://github.com/Keesan12/martin-loop/tree/df388508ac1451a7c3feb12b86ee424c60566208
- Locator: grep -rIlE -i 'steer|send.*message.*running|stdin\.write' packages/cli/src packages/core/src packages/mcp/src packages/adapters/src packages/contracts/src plugins scripts (tests, fixtures and lockfiles excluded)
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-17
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages/cli/src

### martinloop:workspaces

Swarm workspaces. Swarm child, integration and verifier workspaces are Git worktrees (or independent clones) created at a baseline commit and removed after evidence persists.

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/packages/core/src/swarm/workspaces.ts
- Locator: line 260 `worktree add --detach`; isolationMode
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-16
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages/core/src/swarm/workspaces.ts

### martinloop:extout

External outcome verification. GET-only JSON read-back verification of external actions; --approve-external-writes and single attempt for staging_controlled profile.

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/docs/reference/external-outcomes.md
- Locator: Standalone; Governed external-side-effect run
- Evidence kind: documented
- Source date: Not stated
- Retrieved: 2026-11-16

### martinloop:rev-srch-cost

Reviewer search: compute cost. Reviewer bounded search at pinned commit: no hits.

- URL: https://github.com/Keesan12/martin-loop/tree/df388508ac1451a7c3feb12b86ee424c60566208
- Locator: grep -rIlE -i 'computeCost|compute fee|sandbox hours' packages/cli/src packages/core/src packages/mcp/src packages/adapters/src packages/contracts/src plugins scripts (tests, fixtures and lockfiles excluded)
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-17
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages/cli/src

### martinloop:rev-srch-epic

Reviewer search: epics. Reviewer bounded search at pinned commit: no hits.

- URL: https://github.com/Keesan12/martin-loop/tree/df388508ac1451a7c3feb12b86ee424c60566208
- Locator: grep -rIlE -i '\bepic\b' packages/cli/src packages/core/src packages/mcp/src packages/adapters/src packages/contracts/src plugins scripts (tests, fixtures and lockfiles excluded)
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-17
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages/cli/src

### martinloop:rev-sched-fair

Swarm scheduler budget leases. The only "fair-share" wording concerns initial budget lease reservations for sibling children, reclaimed when a child needs more capacity; slot and task ordering carry no documented fairness policy.

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/packages/core/src/swarm/scheduler.ts
- Locator: scheduler.ts lines 150-200 (reserveSwarmBudgetLease, extendSwarmBudgetLease and its fair-share comment); lines 419-470 (concurrency limit)
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-17
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages/core/src/swarm/scheduler.ts

### martinloop:rev-srch-media

Reviewer search: media capture and previews. Reviewer bounded search at pinned commit: 1 file hit(s), reviewed: packages/cli/src/audit.ts. The one hit is a verifier-command detection regex in audit.ts line 7 (mentions playwright test), not capture or preview code.

- URL: https://github.com/Keesan12/martin-loop/tree/df388508ac1451a7c3feb12b86ee424c60566208
- Locator: grep -rIlE -i 'screencast|screen record|playwright|puppeteer|gallery|preview (url|server|app)' packages/cli/src packages/core/src packages/mcp/src packages/adapters/src packages/contracts/src plugins scripts (tests, fixtures and lockfiles excluded)
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-17
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages/cli/src

### martinloop:srch-merge

Bounded source search. No merge, rebase or deploy code; only `gh pr create` and `git reset --soft` for rollback.

- URL: https://github.com/Keesan12/martin-loop/tree/df388508ac1451a7c3feb12b86ee424c60566208/packages
- Locator: rg -i "gh pr merge|pr merge|rebase|cherry-pick" packages/*/src
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-16
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages

### martinloop:rev-srch-group

Reviewer search: grouping. Reviewer bounded search at pinned commit: no hits.

- URL: https://github.com/Keesan12/martin-loop/tree/df388508ac1451a7c3feb12b86ee424c60566208
- Locator: grep -rIlE -i 'group runs|groupBy|group-by' packages/cli/src packages/core/src packages/mcp/src packages/adapters/src packages/contracts/src plugins scripts (tests, fixtures and lockfiles excluded)
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-17
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages/cli/src

### martinloop:repoapi

GitHub repository metadata. 286 stars, Apache-2.0, pushed 2026-11-04, homepage martinloop.com; latest release v0.9.0 (2026-11-03).

- URL: https://api.github.com/repos/Keesan12/martin-loop
- Locator: stargazers_count, license, pushed_at, homepage
- Evidence kind: documented
- Source date: Not stated
- Retrieved: 2026-11-16

### martinloop:telemetry

CLI telemetry. Anonymous allowlisted events after a one-time non-blocking disclosure; disabled by `martin telemetry off`, DO_NOT_TRACK, MARTIN_TELEMETRY_DISABLED or CI.

- URL: https://github.com/Keesan12/martin-loop/blob/df388508ac1451a7c3feb12b86ee424c60566208/packages/cli/src/telemetry.ts
- Locator: lines 20-140
- Evidence kind: source-inspected
- Source date: Not stated
- Retrieved: 2026-11-16
- Commit: df388508ac1451a7c3feb12b86ee424c60566208
- Path: packages/cli/src/telemetry.ts
