35 of 51 criteria have a sourced finding; 25 were also cross-checked by a second agent. Every comparison that includes ai-devkit links here for scope, limitations, atomic answers and sources.
Humans approve the rubric, templates and prototype; they do not review every finding. An agent source cross-check is not a hands-on product test. A missing sourced finding is not evidence that a capability is absent.
AI DevKit has native macOS/Linux CLI/TUI operation; the managed console is unsupported on Windows. The scoped clients are terminal interfaces, with separate Telegram/Slack bridges rather than product web or phone apps.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
Not separately cross-checked
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
9 atomic answers
Desktop os [partial] macOS/Linux managed terminal interface; Windows console unsupported. Sources: ai-devkit:gap35-clients.
Mobile apps [unsupported] No native iOS/Android client in the scoped application build targets. Sources: ai-devkit:gap35-clients.
Browser [unsupported] The scoped product contains CLI/TUI and documentation website, without a browser control client. Sources: ai-devkit:gap35-clients.
Mobile browser [unsupported] Chat-service bridges do not provide a product-owned phone-browser client; no such client is built in the scoped tree. Sources: ai-devkit:gap35-clients.
Managed start supports named coding harnesses and supervision across providers; configuration integration alone is not proof that every editor is launchable. AI DevKit cli@0.73.1 registers AntigravityCliAdapter and a startable antigravity_cli runtime whose command is agy. The complete startable runtime registry omits Cursor and Aider.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 Antigravity source pass: cli@0.73.1 (4a8e19d00ff770ac821f5635144e7d3a12ebcebe). · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
Not separately cross-checked
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
Antigravity CLI must be installed and authenticated; editor configuration support alone does not establish a startable agent.
5 atomic answers
Harnesses [supported] Claude Code, Codex, Gemini CLI, OpenCode, Pi, Grok CLI, Copilot, Kiro, Devin, Muse and Antigravity CLI. AI DevKit cli@0.73.1 registers AntigravityCliAdapter and a startable antigravity_cli runtime whose command is agy. Sources: ai-devkit:console, ai-devkit:readme, ai-devkit:agy-release-1, ai-devkit:agy-release-2.
Model providers [unknown] The checked first-party sources do not establish model providers for this scoped product.
Custom adapters [partial] Managed start selects the shipped fixed runtime profiles; Cursor/Aider are not startable profiles in this revision. Sources: ai-devkit:gap35-harness-registry.
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-08
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-08
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
CLI exposes managed agent control, memory and daemon operations; memory MCP server provides tool access.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-08
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-08
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
7 atomic answers
Mcp client [unknown] The checked first-party sources do not establish mcp client for this scoped product.
Lifecycle skill writes requirements/design/planning/implementation/testing files under docs/ai and requires staged review by instruction.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-08
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
Local SQLite memory supports titled/tagged/scope records through MCP and CLI, retrieved when relevant.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-08
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
10 atomic answers
Instructions [unknown] The checked first-party sources do not establish instructions for this scoped product.
Skills [supported] memory skill for agents without MCP. Sources: ai-devkit:memory.
Project docs [unknown] The checked first-party sources do not establish project docs for this scoped product.
Memory [supported] SQLite decisions, conventions and fixes. Sources: ai-devkit:memory.
Scope [supported] Global and repo/project memory scope. Sources: ai-devkit:memory.
Provenance [unknown] The checked first-party sources do not establish provenance for this scoped product.
Correction [unknown] The checked first-party sources do not establish correction for this scoped product.
Deletion [unknown] The checked first-party sources do not establish deletion for this scoped product.
Freshness [unknown] The checked first-party sources do not establish freshness for this scoped product.
Takeover continuity [unknown] The checked first-party sources do not establish takeover continuity for this scoped product.
Memory retrieval supplies prior project knowledge on demand. This is not a full code index.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-08
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
10 atomic answers
Code search [unknown] The checked first-party sources do not establish code search for this scoped product.
Indexing [unknown] The checked first-party sources do not establish indexing for this scoped product.
External knowledge [partial] Project memory retrieval via search. Sources: ai-devkit:memory.
Multimodal inputs [unknown] The checked first-party sources do not establish multimodal inputs for this scoped product.
Freshness [unknown] The checked first-party sources do not establish freshness for this scoped product.
Access scope [unknown] The checked first-party sources do not establish access scope for this scoped product.
Api retrieval [unknown] The checked first-party sources do not establish api retrieval for this scoped product.
Citations [unknown] The checked first-party sources do not establish citations for this scoped product.
Conflict detection [unknown] The checked first-party sources do not establish conflict detection for this scoped product.
Review enforcement [unknown] The checked first-party sources do not establish review enforcement for this scoped product.
The optional first-party task-manager plugin provides persistent independent tasks, statuses, phases, blockers, evidence, artifacts, ownership and event history through ai-devkit task.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
official-integration
Checked
2026-11-13 (9 days old)
Agent cross-check
Not separately cross-checked
Source inspection covers the scoped release only. Unanswered atomic questions remain unknown; no live scenario was run.
Lifecycle skill persists requirements/design/plans in repo Markdown; no server-enforced spec-drift gate established.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-08
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
Several agents can run in separate managed sessions and be supervised together.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-08
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
Lead uses named workers and start-then-assign handshake; same-folder session detection collisions are documented.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-08
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
7 atomic answers
Human assignment [unknown] The checked first-party sources do not establish human assignment for this scoped product.
AI DevKit orchestration skills equip a lead to start, message, inspect and wait for workers across harnesses.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-08
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
The scoped CLI/daemon and official plugin set has no schedule or external-event service that starts agent runs. A daemon discovery sweep only refreshes session state.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-13 (9 days old)
Agent cross-check
Not separately cross-checked
Source inspection covers the scoped release only. Unanswered atomic questions remain unknown; no live scenario was run.
8 atomic answers
Schedules [unsupported] No product schedule registration/start service in the complete scoped command/RPC/plugin set. Sources: ai-devkit:gap35-control-boundaries.
Events [unsupported] Chat bridge messages steer an already connected agent; no external-event launcher is implemented. Sources: ai-devkit:gap35-control-boundaries.
Recurring tasks [unknown] The checked first-party sources do not establish recurring tasks for this scoped product.
Completion conditions [unknown] The checked first-party sources do not establish completion conditions for this scoped product.
Scheduler location [unknown] The checked first-party sources do not establish scheduler location for this scoped product.
Overlaps [unknown] The checked first-party sources do not establish overlaps for this scoped product.
Retries [unknown] The checked first-party sources do not establish retries for this scoped product.
Client independence [unknown] The checked first-party sources do not establish client independence for this scoped product.
Console messages or Telegram/Slack bridge can steer running agents; managed kill requires confirmation.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-08
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
10 atomic answers
Pause [unknown] The checked first-party sources do not establish pause for this scoped product.
Composable lifecycle/verification/orchestration skills guide workflows without prescribing one fixed agent.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-08
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
10 atomic answers
Quick task [unknown] The checked first-party sources do not establish quick task for this scoped product.
Growing epic [unknown] The checked first-party sources do not establish growing epic for this scoped product.
The local CLI/daemon runs agents on its own macOS/Linux host, including a user-controlled always-on Linux server reached by SSH. systemd user persistence survives logout; this does not keep a sleeping execution host running.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
Not separately cross-checked
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
Byo remote [supported] Install the same core CLI/daemon on a user-controlled Linux host; its systemd persistence and tmux execution survive a controlling SSH client logout. Sources: ai-devkit:gap35-runtime.
Self hosted [unknown] The checked first-party sources do not establish self hosted for this scoped product.
Managed cloud [unsupported] README explicitly not a hosted service and no remote cloud agent execution. Sources: ai-devkit:readme, ai-devkit:runtime.
Hybrid [unknown] The checked first-party sources do not establish hybrid for this scoped product.
Provisioning [unknown] The checked first-party sources do not establish provisioning for this scoped product.
Payer [unknown] The checked first-party sources do not establish payer for this scoped product.
Resources [unknown] The checked first-party sources do not establish resources for this scoped product.
Region [unknown] The checked first-party sources do not establish region for this scoped product.
Expiry [unknown] The checked first-party sources do not establish expiry for this scoped product.
Client disconnect [supported] Daemon/systemd and tmux sessions remain on the execution host when the remote control client disconnects, provided that host remains running. Sources: ai-devkit:gap35-runtime.
Other device continuation [unknown] The checked first-party sources do not establish other device continuation for this scoped product.
Worktree setup is manual guidance: users create directories and pass --cwd. The scoped runtime bundle supplies terminal processes without product-owned container or VM isolation.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
Not separately cross-checked
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
12 atomic answers
Checkout [unknown] The checked first-party sources do not establish checkout for this scoped product.
Branch [unknown] The checked first-party sources do not establish branch for this scoped product.
Worktree [partial] User runs git worktree add manually and supplies the directory; the ADE does not provision the worktree itself. Sources: ai-devkit:gap35-runtime.
Container [unsupported] Complete runtime bundle is tmux/Herdr without a product container sandbox backend. Sources: ai-devkit:gap35-runtime.
Vm [unsupported] No VM/microVM isolation backend in the scoped runtime bundle. Sources: ai-devkit:gap35-runtime.
Os sandbox [unknown] The checked first-party sources do not establish os sandbox for this scoped product.
Filesystem [unknown] The checked first-party sources do not establish filesystem for this scoped product.
Process [unknown] The checked first-party sources do not establish process for this scoped product.
Network [unknown] The checked first-party sources do not establish network for this scoped product.
Secrets [unknown] The checked first-party sources do not establish secrets for this scoped product.
Code separation [unknown] The checked first-party sources do not establish code separation for this scoped product.
Managed durable mode is limited to Claude/Codex/Pi; daemon persistence can be installed on Linux. No full worker crash/reboot recovery test was run.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-08
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
12 atomic answers
Client disconnect [unknown] The checked first-party sources do not establish client disconnect for this scoped product.
Worker crash [unknown] The checked first-party sources do not establish worker crash for this scoped product.
Service restart [partial] Linux systemd user unit persists daemon, not proof all worker processes survive. Sources: ai-devkit:console, ai-devkit:runtime.
Expiry [unknown] The checked first-party sources do not establish expiry for this scoped product.
send --wait supports timeout and readiness handshake; quotas are readings, not enforced spend limits.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-08
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
8 atomic answers
Time limits [unknown] The checked first-party sources do not establish time limits for this scoped product.
Turn limits [unknown] The checked first-party sources do not establish turn limits for this scoped product.
Token limits [unknown] The checked first-party sources do not establish token limits for this scoped product.
Spend limits [unknown] The checked first-party sources do not establish spend limits for this scoped product.
Console shows last-observed running/waiting/idle/unknown status and transcripts; status readiness is explicitly a hint.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-08
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
Capacity reports provider quota windows and reset times. It does not establish per-task billed cost.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-08
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
14 atomic answers
Task cost [unknown] The checked first-party sources do not establish task cost for this scoped product.
Session cost [unknown] The checked first-party sources do not establish session cost for this scoped product.
Child cost [unknown] The checked first-party sources do not establish child cost for this scoped product.
Epic cost [unknown] The checked first-party sources do not establish epic cost for this scoped product.
Harness aggregation [unknown] The checked first-party sources do not establish harness aggregation for this scoped product.
Billed vs estimated [partial] Quota availability is distinct from per-task invoice. Sources: ai-devkit:capacity.
Tokens [unknown] The checked first-party sources do not establish tokens for this scoped product.
Subscription allowance [supported] Provider quota windows read from existing credentials. Sources: ai-devkit:capacity.
Provider reconciliation [unknown] The checked first-party sources do not establish provider reconciliation for this scoped product.
Compute cost [unknown] The checked first-party sources do not establish compute cost for this scoped product.
Refresh delay [unknown] The checked first-party sources do not establish refresh delay for this scoped product.
Alerts [unknown] The checked first-party sources do not establish alerts for this scoped product.
Export [unknown] The checked first-party sources do not establish export for this scoped product.
Drilldown [unknown] The checked first-party sources do not establish drilldown for this scoped product.
Slack proof-of-concept bridge permits multiple workspace members to send text toward one agent; Telegram allows only first user per bridge session.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
Not separately cross-checked
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
Discussion [unknown] The checked first-party sources do not establish discussion for this scoped product.
Comments [unknown] The checked first-party sources do not establish comments for this scoped product.
Presence [unknown] The checked first-party sources do not establish presence for this scoped product.
Coprompting [partial] Any workspace member who can DM Slack app can send to connected agent; response destination first DM only. Sources: ai-devkit:channel.
Ordering [unknown] The checked first-party sources do not establish ordering for this scoped product.
Conflicts [unknown] The checked first-party sources do not establish conflicts for this scoped product.
Attribution [unsupported] Responses route to the first DM until restart and are not attributed per person; bounded source search found no attribution or sharing feature. Sources: ai-devkit:d80-channel, ai-devkit:d105-collab.
Permissions [partial] Slack user authorization not enabled; use disposable private workspace. Sources: ai-devkit:channel.
Plan restrictions [unknown] The checked first-party sources do not establish plan restrictions for this scoped product.
Owner online [unknown] The checked first-party sources do not establish owner online for this scoped product.
The scoped application uses local OS-account state and chat-service bridge authorization. It has no named human workspace roles or enterprise SSO.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-13 (9 days old)
Agent cross-check
Not separately cross-checked
Source inspection covers the scoped release only. Unanswered atomic questions remain unknown; no live scenario was run.
9 atomic answers
Organization roles [unsupported] No human workspace/organization role or membership model in the scoped app. Sources: ai-devkit:gap35-control-boundaries.
Project roles [unknown] The checked first-party sources do not establish project roles for this scoped product.
Task roles [unknown] The checked first-party sources do not establish task roles for this scoped product.
Membership [unknown] The checked first-party sources do not establish membership for this scoped product.
Underlying harness permissions/trust must be configured; tested presets are Linux 0.70.1 guidance with broad access.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-08
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
14 atomic answers
Command [unknown] The checked first-party sources do not establish command for this scoped product.
Phase Markdown and historical transcripts remain available locally; output can be piped into agent send. The product console has no dedicated code-change diff review view.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
Not separately cross-checked
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
13 atomic answers
Diffs [unsupported] Console renders transcript Markdown and memory; no product-owned proposed-code diff viewer in the scoped client. Sources: ai-devkit:gap35-control-boundaries.
verify/TDD/review skills instruct fresh verification before completion; no independent runtime-enforced gate was established.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-08
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
Shipped guides can instruct an agent to use Git and manually created worktrees. The scoped CLI/plugin implementation has no product-owned PR creation or management workflow.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-13 (9 days old)
Agent cross-check
Not separately cross-checked
Source inspection covers the scoped release only. Unanswered atomic questions remain unknown; no live scenario was run.
9 atomic answers
Branches [unknown] The checked first-party sources do not establish branches for this scoped product.
Worktrees [unknown] The checked first-party sources do not establish worktrees for this scoped product.
Commits [unknown] The checked first-party sources do not establish commits for this scoped product.
Pull requests [unsupported] Task branch/PR links are metadata; no PR creation/management CLI or RPC is supplied. Sources: ai-devkit:gap35-control-boundaries.
Comments [unknown] The checked first-party sources do not establish comments for this scoped product.
Ci response [unknown] The checked first-party sources do not establish ci response for this scoped product.
Rebase [unknown] The checked first-party sources do not establish rebase for this scoped product.
Conflicts [unknown] The checked first-party sources do not establish conflicts for this scoped product.
Product state [unknown] The checked first-party sources do not establish product state for this scoped product.
AI DevKit application source is Apache-2.0 licensed.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-08
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
Core CLI/daemon, memory and console run on user-controlled machine; no hosted service required.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-08
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
README explicitly says no telemetry and local memory. Channel tokens are local plaintext files mode 0600; agent/provider terms remain separate.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-17
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
Telemetry defaults [supported] No telemetry is collected: the README says no telemetry and bounded source searches found no collector. Sources: ai-devkit:d80-readme, ai-devkit:d80-negtree.
Telemetry controls [unknown] The checked first-party sources do not establish telemetry controls for this scoped product.
Consent prompt [unsupported] No telemetry consent prompt exists because no telemetry exists. Sources: ai-devkit:d80-negtree.
AI DevKit is free (Apache-2.0) and not a hosted service; no paid plan found, and the agents/models it coordinates are paid separately to their providers. The managed Gemini CLI worker can use a user-supplied Google API key, so provider model usage can be paid through the user’s own key.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: First-party pricing sources read 2026-11-10 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-10 (12 days old)
Agent cross-check
2026-11-17
First-party pricing pages, terms and licences read 2026-11-10; no purchase, billing or authenticated account test.
Provider credentials and model usage are separate from the free ADE application; no billing test ran.
12 atomic answers
Currency [unknown] Not stated in the sources checked (homepage, README, LICENSE).
Interval [unknown] Not stated in the sources checked (homepage, README, LICENSE).
Seats [unknown] Not stated in the sources checked (homepage, README, LICENSE).
Seat minimum [unknown] Not stated in the sources checked (homepage, README, LICENSE).
Workspace fees [unsupported] No fees: the site states free and Apache-2.0 licensed and has no pricing page. Sources: ai-devkit:d80-site, ai-devkit:d80-readme.
Included usage [unknown] Not stated in the sources checked (homepage, README, LICENSE).
Trial [unsupported] No trial: free software. Sources: ai-devkit:d80-site.
Overages [unknown] Not stated in the sources checked (homepage, README, LICENSE).
Taxes [unknown] Not stated in the sources checked (homepage, README, LICENSE).
Model cost [partial] The official per-harness setup guide explicitly supports a user-supplied Google API key for managed Gemini CLI workers. This is a provider BYOK model-use path, separate from the free ADE application. Provider rates are separate from the ADE fee. Sources: ai-devkit:gap35-provider-keys.
Compute cost [partial] Runs locally and does not run a remote cloud agent, so there is no vendor compute charge described. Sources: ai-devkit:readme-nov10.
Latest listed CLI release 0.72.0 was published November 3; console remains experimental macOS/Linux and unsupported Windows.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-08 (14 days old)
Agent cross-check
2026-11-08
Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
AI DevKit is free Apache-2.0 software run locally through npx, with no paid plan or hosted service. The coding agents it supervises are paid to their providers.
Scope
Surfaces: AI DevKit CLI, experimental agent console and devkitd · Version: Source documentation at 99ac1c1864749daa5b1387ad16215ff9b0eb08d9 · Plan: Not specified · OS: Not specified · Execution modes: local · Region: Not specified · Harness: Not specified
Availability
generally-available
Delivery
native
Checked
2026-11-10 (12 days old)
Agent cross-check
2026-11-10
Free scenarios, usage caps and card requirements for the underlying agents were not checked. No installed-product test was run.
Local-first control layer offers unified configuration, console, cross-agent messages, local SQLite memory and composable lifecycle/verification skills. It explicitly is not a hosted service, has no telemetry and does not run remote cloud agents.
Kind
documented
Locator
Start in 30 seconds; Operate agents; Memory; Workflows; Boundaries
Product orchestration skill guides a lead using agent start/send/list to supervise named workers and cross-harness review. Users manually supply separate worktree directories; same-folder session matching may collide. Console experimental on macOS/Linux and unsupported Windows.
Kind
documented
Locator
Lead vs console; Layout; Skills; Start-then-assign; Known limitations
Managed agent start and experimental terminal console support macOS/Linux and explicitly do not support Windows. The scoped product client tree contains CLI/TUI, without browser, native iOS/Android or phone-browser control clients; its web directory is a documentation site and chat-service bridges are separate.
Kind
documented
Locator
Prerequisites; packages/cli client tree; web documentation routes; release/package targets
The complete startable runtime registry includes Claude, Codex, Copilot, Gemini CLI, Grok CLI, Antigravity CLI, OpenCode, Pi, Kiro, Devin and Muse. Cursor and Aider have no startable runtime profile or coding-harness adapter; configuring editor files is separate.
Readiness checks are hints, not successful login probes. Capacity reads stored provider credentials and reports quota windows, availability and reset times; it does not establish task billing.
Local SQLite stores titled/tagged project/global memory. Memory MCP server and CLI provide retrieval/storage; setup wires compatible agents, Pi uses CLI/skill instead.
The official installable task-manager plugin adds ai-devkit task create/list/show/update/status/progress/assign/note/close commands. Stable independent task IDs, metadata and append-only events persist in tasks.db, separate from agent sessions.
Kind
documented
Locator
CLI and Storage; src/command.ts; src/database/migrations/001_initial.sql tasks and task_events
The scoped daemon serves local Unix-socket agent RPCs; CLI commands and shipped plugins contain no schedule/webhook service that starts work, human membership/role or enterprise SSO model. Console preview renders conversation Markdown rather than a proposed-code diff viewer; Git/PR references in task records are metadata, without PR creation/management commands.
Bridge forwards messages to running terminal agents and polls output. Telegram restricts each session to first user. Slack proof of concept allows workspace members who can DM the app to send to same connected agent; responses route to first DM until restart.
The complete runtime bundle launches agents through tmux or Herdr on the daemon host. The local Unix-socket control plane can run on a user-controlled Linux server and systemd keeps it alive after logout; no vendor service is required. It has no automatic worktree provisioning, container isolation or VM runtime.
Kind
source-inspected
Locator
Runtimes and StartInput; runtime/tmux.rs and herdr.rs; web/content/docs/16-runtime.md Linux persistence; orchestration guide manual git worktree setup
Linux 0.70.1 guidance configures underlying harness allow-lists/trust, with broad shell/file permissions. Other harnesses need verification. This is not an AI DevKit security sandbox.
The Slack bridge is a proof of concept: any workspace member who can DM the app can send to the connected agent and channel messages are unsupported; Telegram restricts a session to the first user.
Bounded searches for telemetry, analytics, posthog, sentry, mixpanel, amplitude and plausible found no collector or toggle (matches are unrelated identifiers and a template mentioning user analytics).
Kind
source-inspected
Locator
packages; case-insensitive searches over packages, web, rust, hooks, npm, templates, skills and docs; cli@0.76.1 (c6f1a0a)
The Slack bridge accepts any DM from the configured workspace (team id, im channel type, not the bot) with no per-user allowlist or attribution; the collaboration-term searches found no sharing, presence, comment or per-person permission feature (only unrelated session-locator identifiers).
Kind
source-inspected
Locator
packages/channel-connector/src/adapters/slack/SlackAdapter.ts l196-l210; case-insensitive searches over packages, web/content and rust for collaborat, multi-user, co-edit, teammate, shared session/agent, presence; cli@0.76.1 (c6f1a0a)
The official ai-devkit orchestration guide explicitly lists Google API-key authentication as an alternative to login for the managed Gemini CLI worker. The application orchestrates agents authenticated to users’ own providers; it does not supply a bundled model quota.
AI DevKit is run free with npx ai-devkit@latest setup and init; the README says it is Apache-2.0, runs locally and is not a hosted service, and names no price or paid plan.
Kind
documented
Locator
README.md setup commands (lines 18-24) and 'Not a hosted service' (line 249), commit 6ddfed6