A source-linked comparison across the complete Podium ADE rubric. 47 criteria sourced for both products; 37 criteria also agent cross-checked for both. 51 rows have at least one sourced finding, including 50 with an agent-cross-checked finding. No score or winner is inferred.
Each row shows a finding's verification status, support, check date and answer. Its scope, limitations, atomic answers and sources are on the product's page. 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.
The Copilot desktop app supports macOS, Linux and Windows, including optional WSL2 execution on Windows. GitHub Mobile provides native Android and iOS clients with Copilot chat and issue delegation. Copilot in the cloud starts from GitHub.com; GitHub Mobile is also an entry point. The CLI supports Linux, macOS and Windows through PowerShell or WSL. Copilot Chat is documented in IDEs; per-editor feature parity remains unverified. The app starts and steers sessions, reviews changes and manages PRs. Copilot in the cloud is available in a mobile browser, including use of connected tools. GitHub Mobile live notifications cover cloud agent and remote CLI sessions with In progress, Waiting for user input, Idle and Finished states. Desktop app notifications are not documented in these sources. Cloud sandbox sessions continue on any client and every surface shows the same session list; a stopped sandbox resumes from a snapshot; app v1.2.0 adds a GitHub Mobile QR code to continue cloud sessions on a phone.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Sourcedpartial
Podium ships desktop, CLI and browser clients, including phone-browser control. The native iPhone app is private beta; the complete download offering contains no native Android app.
GitHub supports public-preview Anthropic Claude using Claude Agent SDK and OpenAI Codex using Codex SDK alongside Copilot. Users assign a prompt or issue to a partner coding agent, receive a PR and request iteration through comments. Generally available models come from OpenAI, Anthropic, Google, Microsoft, Moonshot AI and xAI. User BYOK connects OpenAI-compatible endpoints, Azure OpenAI, Anthropic and local endpoints such as Ollama or vLLM. Adapters for other agent harnesses are not documented. The Agents tab starts sessions with Copilot, Anthropic Claude or OpenAI Codex (Claude and Codex in public preview). Each is a separate agent session; combining harnesses within one task is not documented.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Podium launches the native Claude Code, Codex, Grok, OpenCode, Cursor and Pi CLIs in persistent terminals on machines you control. The 0.2.0 release calls Pi fully supported. Models come from each harness catalog, and capabilities differ per harness: Pi declares no quota, handoff or hooks. Mixed-harness operation was not exercised.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
Partner Claude and Codex agents can use the existing Copilot subscription; this does not establish use of separate Claude or ChatGPT consumer subscriptions. Current plans allocate GitHub AI credits; organization usage can draw from a shared pool. Partner agents require a paid Copilot plan and separately enabled account/organization policies; integrations are public preview. User-configured BYOK works in VS Code, JetBrains, Xcode, Copilot CLI, the app and the SDK; requests go directly to the provider. GitHub.com is not a listed BYOK client. BYOK can use Ollama and other local or on-network endpoints, including in air-gapped environments when the endpoint is reachable. The app signs in to github.com or GHE.com and, since v1.2.0, can add GitHub Enterprise Server accounts in Settings > Accounts. Switching between several github.com Copilot accounts is not documented. BYOK works without a Copilot subscription, but the app still requires a GitHub account and sign-in.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Podium supports native coding-agent logins and managed provider credentials. Claude subscriptions (via native login or a managed setup-token) and local ChatGPT/Codex login are supported paths, alongside Anthropic and OpenAI API keys. Docs list billing and hosted metering as not implemented product claims.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
Cloud work requires repositories hosted on GitHub; other Git hosting is explicitly excluded for this workflow. Cloud sandboxes and GitHub Actions-based environments have different setup and customization controls. Prepare a Linux host, authenticate/register its environment, and select it from a client signed into the same GitHub account. Documented app path: install, sign in to GitHub, choose a plan or model provider, connect a repository or folder, pick a session mode, prompt, review Changes, then Create PR. Remote control accepts CLI sessions in directories not associated with a repository, and the app can connect a local folder. The app lists Git as a prerequisite; full non-Git operation is not documented. Remote control supports non-GitHub repositories and the app clones repositories from other Git hosts by URL. Cloud agent sessions only work with GitHub-hosted repositories. App onboarding offers a Copilot plan or a BYOK model provider; without write access a cloud session is read-only. Error guidance for missing prerequisites is not documented.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Sourcedpartial
Docs describe a first-run flow (install, podium, register a Git repository, confirm a detected logged-in harness) and a headless VPS installer with machine joining. Setup time and failure rate were not measured, and no self-serve managed cloud setup is documented. The current README and installation guide use podium.do/download and podium.do/install.sh as the canonical stable entrypoints.
copilot -p runs a single prompt and exits; tool approval flags scope execution. The agent tasks API creates, lists and gets cloud agent tasks; REST and GraphQL issue APIs assign issues to Copilot. Both are public preview; stop and steer endpoints are not documented. The Copilot SDK client library (MIT; moved to microsoft/copilot-sdk as Microsoft Copilot SDK) can start sessions, including cloud sandbox sessions. The MIT licence covers only the SDK client: SDK use needs a Copilot subscription unless BYOK keys are configured, and cloud sandbox sessions run on GitHub's hosted service under Copilot plan terms and billing. The API creates, lists and reads tasks; CLI and remote control can steer, stop and approve permission requests. The tasks API accepts only user-to-server tokens (PAT, OAuth app or GitHub App user token). GitHub App installation tokens are supported only for SDK server-to-server use, in public preview. The SDK is generally available and follows semantic versioning; the latest stable release is v2.0.1 (2026-11-18), and prereleases are also published (v2.0.2-preview.4 on 2026-11-21). The tasks API, the issue-assignment API and SDK server-to-server authentication are public preview.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
The podium CLI covers issue create, show, update, claim, dependencies, artifacts, session send/status/read/stop and review actions, and the server exposes the same issue commands as a bearer-token MCP tool endpoint. No documented public HTTP API, SDK or webhooks were found.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
Cloud work supports MCP; personal MCP and connectors in cloud sandboxes differ from repository MCP on Actions. The app Customize area discovers and manages plugins, skills, MCP servers and canvases; the app can install MCP servers from Customize. Custom agent profiles can restrict available tools; hooks run commands during agent work. Personal, repository and organization custom instructions and agent skills apply; Copilot CLI imports prepared memories in bulk. GitHub handles authentication for connectors; personal MCP servers support only key-based auth via COPILOT_MCP_ secrets; repository and custom-agent MCP servers do not support remote OAuth servers. The default GitHub MCP token is read-only and scoped to the current repository; the tools field allowlists tools; delegated runs call configured MCP tools without approval. Copilot owns the integration layer: GitHub-managed connectors, administrator-configured repository MCP servers and user-configured personal MCP servers.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Podium gives agents its own issue tools over CLI and MCP, and agent profiles import and export Claude Code agent files. No Podium documentation for user-configured MCP clients, plugins or custom tools was found.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-07. Scope and sources
Official docs state only that the app is built on GitHub Copilot CLI; the desktop UI framework is not documented.
Checked 2026-11-21 (1 day old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Desktop releases use a Tauri 2 shell that serves the Podium web UI; mobile source uses Expo and React Native. A native iPhone release is described as private beta.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
The GitHub.com chat/agent flow accepts repository-context prompts. Work can originate in GitHub issues; GitHub remains the issue system. Sessions start from prompts on web, app, CLI, IDE and Mobile; @GitHub in Slack and Teams creates issues. GitHub issues can be assigned to Copilot; Linear and Sentry connectors provide context only. @copilot on PRs, @GitHub in Slack and Teams (GA 2026-11-06) and automation triggers on issue and PR events start work. GitHub issue numbers and URLs; agent tasks carry task IDs. GitHub Issues is the native tracker, so agents read and assign issues on the same platform without a sync layer. Sync with external trackers such as Linear is not documented; connectors provide context. Automations can be granted tools to label, comment, mark PRs ready or draft and merge PRs.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Podium issue records are native and authoritative with stable references, CLI creation and editing, and agent-driven filing from conversation. The in-product assistant has Linear search, create and move tools when a key is configured, but no import or two-way tracker sync was found.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
CLI plan mode asks clarifying questions and builds an implementation plan before code changes. Project Overview relates sessions to their plan, code changes and PR. Humans can give a project follow-up work and direction. Issue-artifact Markdown is editable on github.com (public preview); app canvases let people and agents work on the same plan. Issue artifact edits keep version history (public preview). Stopped cloud sandbox sessions keep a snapshot and resume from any client.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Issue brief, acceptance criteria, state paragraph and todos persist on the durable issue record. Release 0.3.0 removed the separate design and notes fields and appended their text to the brief. On the human issue page the title, description and acceptance are editable and the brief is read-only; agents edit the brief and other fields through issue commands.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-11. Scope and sources
Parent issues hold sub-issues; projects group by parent issue. Up to eight nesting levels and 100 sub-issues per parent, including issues from other repositories. Blocked-by and blocking relationships are set in the issue UI and the gh CLI. An existing issue can become a parent by adding sub-issues; linkage to session history is not documented. An existing issue can gain sub-issues and keeps its number and discussion; carry-over of session outputs is not documented. No mandatory workflow: chats, single-repository sessions or projects.
Checked 2026-11-21 (1 day old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Podium supports persistent parent and child issues, phases and typed dependency links, including reparenting an existing issue without replacing its identity. Since 0.3.0 issue types are task, phase and decision: epic and other kinds are retired, and any issue with children acts as a parent. Dependency types are fixed to blocked-by, related, discovered-from, supersedes and duplicates. Cycles are refused and docs list an open blocker as a reason an issue cannot start. No mandatory workflow preset applies in the baseline.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
Custom instructions, skills and custom agents apply with environment-specific configuration. Agent skills work on github.com, Mobile, the app and CLI. App projects hold repositories, design mocks, Slack threads and notes (public preview). Copilot Memory stores repository facts and user preferences across sessions and surfaces. Repository facts are per repository (not pooled, not copied to forks); preferences are per user. Repository facts cite code; user preferences cite the user's words. Memories cannot be edited directly on GitHub.com; Copilot upvotes or downvotes them and users can delete them. Write-access users delete repository facts, users delete their preferences, and Business/Enterprise admins can delete preferences. Memories expire after 90 days by default (7-180 configurable for repository facts); citations are checked against the current branch. Repository facts are visible to read-access users and shared across surfaces; colleague takeover is not documented.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Sourcedpartial
v0.3.0 routes Podium issue instructions and relay environment through terminal and engine session launch. Claude stream sessions use appended instructions and boundary callbacks; Codex server sessions use developer instructions, Grok ACP uses rules, and OpenCode uses the prompt system field. Claude and Codex boundary hooks can re-prime after compaction. Transcript mirroring is on by default; full-text indexing needs the command-palette flag, which defaults off. Agent profiles can carry markdown instructions. Curated memory, skills and memory correction or deletion controls remain unestablished. No instruction-delivery or compaction scenario was run.
GitHub chat documents semantic code search limited to default branches. Built-in connectors (Sentry, Linear, Cloudflare), MCP servers and Slack or Figma context. Voice and image input on Mobile; image generation in the app; Playwright screenshots in cloud agent. Memory citations are checked against the current branch. Chat is scoped to one organization or personal repositories by default; the GitHub MCP token is repository-scoped and read-only. Memory facts cite supporting code.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Sourcedpartial
Not established. Podium mirrors daemon transcripts by default, keeps a full-text index only when the command-palette flag is on, and stores session attachments. No repository code search, architecture awareness or API retrieval feature was found.
Copilot drafts GitHub repository issues for human submission. Copilot drafts updates to existing repository issues. Drafted issue fields include labels. Drafted issue fields include assignees. GitHub issues are repository records that can precede an implementation session. Generated issues use repository issue forms/templates and structured fields. Issue and PR comments; only write-access users' comments reach the agent. Project custom fields such as Size can filter issues in the app; a priority field was not checked. The app browses and searches issues. The app filters issues and PRs by custom fields; Mobile filters sessions. Issues close and reopen (both are automation triggers); archiving was not established. Issue reopened is an automation trigger. Projects group issues by parent issue. An issue exists without a running agent until it is assigned to Copilot or a session starts from it. Assign an issue to Copilot or start a session from it. Issues have open, closed and status changes; sessions have task states. No external tracker is needed; GitHub Issues is built in.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Podium exposes native issue creation with an optional start flag, separate dispatch, editing, comments, labels, assignment, search, board filters, archive, sibling ordering and an issue tree. Issue records have their own identity separate from sessions and stages are durable. The idea-to-mission scenario was not run. v0.3.0 refuses an agent close or done update while issue sessions are computing unless --confirm-interrupt is supplied; guidance says to request review and close only on user instruction.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
Copilot can propose issue creation with human draft review. Copilot can propose updates to an existing issue. Issue drafts can include assignees. The user reviews and refines an issue draft before submitting. The same issue numbers and URLs are used through the gh CLI and the GitHub MCP server. The GitHub MCP server reads issues and PRs. Copilot appears as an issue assignee; agent self-claiming is not documented. Granted tools can mark PRs ready or draft and merge them; issue state writes depend on tool scope. The gh CLI edits blocked-by links; agent use of them is not documented. Results are PRs and issue artifacts (public preview); app sessions attach artifacts automatically. Commits are authored by Copilot with the requester as co-author; shared automations act as Copilot; Slack/Teams items carry an attribution line. Write access is required; automation tools can be limited to Only this issue.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
CLI and MCP address the same canonical issue records for reads, creation, edits, claims, state, dependencies and artifacts. Relayed agents have worker capability scoped to their issue subtree, with --outside-scope confirmation for other existing issues. attachSession is not subtree-gated, but direct rehoming from a real issue is refused and an agent attach naming a Proposed target is refused. That does not make every Proposed subtree inert: operator attachment or spawn-time cwd binding can leave a session attached to a Proposed issue; newSubissue without targetId then bypasses the proposed-parent guard, creates a human-audience backlog child and attaches the existing session. This path remains at v0.3.0 source and was not exercised live. Concurrent-edit handling remains unknown.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
No publishable sourced finding is available for this criterion. This does not mean the product lacks the capability.
Podium
Sourcedpartial
The spec CLI and service (living specs under pspec with stable SP IDs) are in v0.3.0. Specs is an experimental, default-off feature listed on the edge channel: the flag controls the Specs UI, the spec line in CLI help and whether agents are told about the spec CLI. The podium spec command itself is dispatched, and the server specs service has no flag check, so direct CLI use does not need the flag. This agrees with the docs, which call specs not baseline rather than blocked. Not run end to end.
The Copilot app runs multiple sessions simultaneously. Independent workstreams can run concurrently. The app sidebar groups agent sessions by repository.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-07. Scope and sources
Podium
Agent cross-checkedpartial
Docs and release notes describe many sessions running as independent terminals, one worktree per issue, and several paired machines under one server, with session and issue state visible on the flight deck, board and CLI. Agent-initiated spawns are limited to 10 per target issue per UTC day. Achieved concurrency is unmeasured.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
The orchestrator assigns context and tools to subagent tasks. Work is assigned through GitHub issue Assignees, including assigning Copilot to an issue. Agent tasks report a queued state; queue management is not documented. Automations route repository events, schedules, API calls and other automation runs to agent runs; Copilot Fusion routes models, not work. Cloud agent can push only to its own copilot/ branch or to the PR branch it was asked to work on; this restricts the agent, not human ownership.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Issues carry an assignee, claim sets assignee and in_progress, a coordinator session can be set per issue, sessions have seats, and work can be placed on a chosen machine or agent profile. claim has no exclusivity check of its own, locks are advisory leases and Podium does not serialize agents editing one worktree, so exclusive ownership is not enforced.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
The main agent is the orchestrator and assigns subagent tasks. Custom agents or models can specialize suitable subtasks. The orchestrator reviews and combines child results. The product-owned Copilot main agent manages child workflow and dependencies. /fleet analyzes task decomposition and starts eligible subagents. /fleet is available in the app and CLI. Partner agents (Claude and Codex, public preview) run as separate sessions; a Copilot coordinator directing them is not documented. Partner agents (public preview) use the user's Copilot sign-in and subscription; enabling them installs a GitHub App per agent. Session logs show progress in real time; per-child progress for coordinated agents is not documented. Partner agent PRs (public preview) accept follow-up requests through PR comments.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Podium orchestrates delegation itself: podium agent spawn starts a full child session on an issue with a chosen harness, model, effort or agent profile, mail carries messages, and bounded await reports the child result. Children are real sessions shown as seats on the flight deck. Agent-initiated spawns take a daily budget per issue. A complete delegated task with retained child findings was not run.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
Automations run on schedules. Repository events and API requests can trigger an automation. Saved automation definitions run repeatedly. Cloud automations run on GitHub infrastructure. A per-run AI credit limit (minimum 30) stops a run; task completion criteria are not documented. Automations run in the cloud and are managed from the Agents tab, the app, VS Code or the REST API.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Sourcedpartial
The server owns a scheduler for cron and one-time automations that spawn agent sessions, skipping overlaps and not backfilling occurrences more than an hour late. The Automations UI is an edge-visibility feature that defaults off and docs list automations as non-baseline. Agents can subscribe to issue and session events.
Steer this session sends further direction from a project. Needs input groups decisions, reviews and permission requests. A stopped cloud sandbox session keeps a snapshot and resumes later; pausing a running task without stopping is not documented. Remote control can stop a session; agent tasks have a cancelled state. A stopped cloud sandbox session resumes with files and work intact from any client. Mobile live notifications include a Waiting for user input state. App projects run agents in the background and check in when they need the user. Tasks report failed and timed_out states; security validation results and actions appear in the session log. Remote control mirrors CLI permission approvals and questions to GitHub Mobile and github.com. Session data is retained 28 days for Actions-based sessions and for the life of the account for cloud sandbox sessions.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Chat and native CLI views share a session. v0.3.0 defaults human controller chat to steer and immediately types it into running Claude Code, Codex, Grok and OpenCode TUIs that declare native busy-input queuing; undeclared harnesses wait for a boundary, and compaction waits. Mail and automation keep when-ready delivery. Stop, resume, questions, decision issues and offers remain available. This is inspected implementation, not a cross-harness runtime parity result.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
A project coordinates related sessions across repositories. Project Overview groups work needing input. Users reopen project sessions, conversations, changes and plans. Blocked issues show a Blocked icon on project boards and Issues pages.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Release 0.3.0 ships a flight deck showing a whole mission in one view: phases, every task and its state, who works on what, what is blocked and by what, and proposals. Done work stays visible. The CLI prints the same snapshot. The waterfall source is dormant and the Timeline view was deleted, so per-task timeline history is not shown. The shared 12-task UX scenario has not been run.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-11. Scope and sources
Chats run without a branch or workspace, and a focused session can start directly in one repository. App projects span repositories and agents (public preview). The app offers Interactive, Plan and Autopilot modes; no preset is required. Automations can propose changes to their own tools, triggers and prompts for the user to apply or dismiss. Model and reasoning effort are set per session; custom agent profiles configure instructions and tools. Users steer cloud sessions mid-run; the app can change sandboxing for an active local session. Business and Enterprise need the app policy (CLI policy also for remote control); partner agents need enabling by policy.
Checked 2026-11-21 (1 day old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
A quick task can be an interactive session or an issue; an issue can grow into a mission by gaining children while keeping its identity; no mandatory workflow preset applies in the baseline. Per-agent harness, model and effort are chosen at spawn or through agent profiles, and the deck shows seats. The coordinator-with-two-reviewers scenario was not run.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
Default project tasks run against the local machine; --cloud selects GitHub-hosted compute instead. Copilot sessions can run on a registered user-managed Linux machine. Cloud sandboxes and Actions-based sessions execute remotely; Actions can use configured self-hosted runners. The user supplies and maintains a Linux execution host and registers it with Copilot. The user is responsible for custom host infrastructure and lifecycle. A cloud sandbox continues working without the client staying connected. Cloud sessions can be continued across supported Copilot clients. Actions-based sessions can run on ARC self-hosted runners; the Copilot service is GitHub-hosted and unavailable for GitHub Enterprise Server. App sessions run locally or in the cloud; cloud sessions can be opened in VS Code or CLI.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Sourcedpartial
The published Podium execution choices are local user machines and a user-owned VPS or paired machine. No vendor-operated agent host is in the complete download offering.
Desktop sessions use Git worktrees and local OS sandbox restrictions. Actions-based cloud-agent sessions can use fresh hosted virtual machines. Current cloud-sandbox documentation does not establish its container or VM implementation. Repositories are cloned into each cloud sandbox. Cloud sandbox commands cannot touch local files; local sandboxing applies path rules. Local sandboxing contains spawned processes; built-in CLI file tools run in-process outside it. Cloud sessions follow the firewall allowlist; local sandboxing has per-host rules. The local proxy swaps credential placeholders for approved hosts; COPILOT_MCP_ secrets serve MCP servers. Local sandboxing is lightweight OS containment, not a VM; trusted-directory scoping is heuristic.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Sourcedpartial
Podium separates issue files with Git branches and worktrees, but its released daemon launches native host processes without product-managed container or VM provisioning.
The user does not need to stay connected while a cloud sandbox works. Returning to a stopped cloud session resumes files and in-progress work. Stopped cloud sandboxes retain a state snapshot. Session logs reopen from the session list. Remote control streams a live CLI session while the machine stays online. The CLI auto-compacts at about 95% of the token limit.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Closing a browser tab does not stop the agent process on a running Podium host. Idle sessions may hibernate under configurable policies; a stopped session keeps its transcript, branch and resume reference. Release 0.2.0 says a failed update rolls back and running sessions keep running through the upgrade; database rollback remains a manual restore. Terminal reattachment goes through the driver adopt contract with process-identity and observation-lease checks. Worker-crash and multi-day behavior were not tested.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
Cloud sandbox state is snapshotted while a session is stopped. Cloud personalizations can use a dotfiles repository; no dotfile install scripts are run. copilot-setup-steps preinstalls tools for Actions sessions; sandboxes copy .copilot dotfiles without running install scripts. The setup steps example caches npm; no general cache feature is described. services is honoured only in Actions setup steps. Agent secrets and variables; COPILOT_MCP_ secrets for MCP servers.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Only machine setup is documented: the installer installs missing runtime tools and pairs a host, and the state directory holds server-managed secrets. Dependency caching, services, ports and automatic repair for project environments were not found.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
Actions-based Copilot sessions stop after a maximum 59 minutes; the limit cannot be extended. Each automation can set AI Credit limits. Per-session AI credit cap in CLI and SDK (soft, public preview); auto-compaction. Tasks have a timed_out state; setup steps timeout is at most 59 minutes. Session limits are a soft cap; budgets can block usage; automation period limits do not stop running runs.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Podium exposes retryable errors, optional auto-continue with backoff, bounded waits and provider quota. Claude stream and hosted headless drivers enforce a supplied turn deadline, with no deadline when timeoutMs is absent. The one-shot headless server runner supplies its budget, default ten minutes. Agent spawns have a daily per-issue budget. Operator-configurable task-wide turn, token and spend caps remain unestablished; quota display is not enforcement.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
Session logs record work and tools used; GitHub says logs do not replace review and testing. OpenTelemetry from CLI, the app and VS Code Agent Host; content capture off by default. AI usage CSV reports and the usage metrics API. The usage metrics API reports distinct and top custom agents, skills and MCP servers per user and per organization; outcome statistics per agent are not documented. Tasks report failed and timed_out states. Tasks report a waiting_for_user state. The Fusion benchmark score is vendor-reported without fixture or attempt detail.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Podium offers local logs and crash export, structured session status, issue stats, a usage and analytics sheet with hourly list-price cost, per-task costs and host load panels. The 0.3.0 notes report a vendor benchmark of about 330 MB server memory against about 1.2 GB for an earlier Podium build, and 8 to 52 percent faster start; the method is published in the repository and was not reproduced here. Usage export and cross-reconnect consistency were not established.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-11. Scope and sources
Each subagent independently consumes model usage and can increase GitHub AI Credit consumption. CLI session limits (public preview) track credits per session with /limits; a per-task cost view on GitHub.com was not found. Session credit tracking with limits (public preview) in CLI and SDK. Billing is per token, but Copilot code review costs are published only as estimated typical ranges. /context shows token usage; billing is per token. AI usage CSV reports. The AI usage CSV includes app_client_id for app-attributed usage.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Since 0.2.0 Podium shows what each task costs, on the task header and the usage sheet, with per-session rows and descendant rollups. The 0.3.0 deck adds a mission cost chip. Figures are token estimates at public model list prices, not subscription invoices or provider bills.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
Agent tasks can be queued. Cloud sandbox memory is billed as allocated; sandbox sizes are not published and CPU or RAM limits for agent tools are not documented. Temporary rate limits tell users to wait and retry. Tasks have a cancelled state; remote control can stop sessions. Cloud sandboxes and GitHub-hosted runners draw from one included compute allowance.
Checked 2026-11-21 (1 day old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Podium reports host memory, load and disk, auto-hibernates idle sessions under pressure, and provides named advisory resource leases. The merge-queue panel was removed; a shipping queue remains behind the Shipping flag. Concurrency limits and bounded test queuing were not established.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
Repository collaborators can follow shared work. Writers can each ask Copilot for changes on the same open PR using @copilot comments. Shared local-session links remain view-only; simultaneous input conflict handling is unestablished. Teams Canvas discussion around plans and changes. PR comments drive follow-up iterations. Slack/Teams attribution line and co-authored commits. Cloud sessions need a paid plan; Free keeps Copilot Chat. Cloud sessions continue without the owner connected; remote-control sessions need the owner's machine online.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
The issue UI displays owner and visibility read-only and deliberately omits sharing controls. Backend share handlers and a Members settings section with invites and admin or member roles exist; they do not establish a finished human collaboration workflow. Docs say small teams can share an instance but the interface prioritizes one operator.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
Enterprise owners set the Copilot in the cloud and third-party agent policies; organization owners block repositories, configure the cloud agent firewall and choose which enabled agents to allow. GitHub documents enterprise roles with permissions (enterprise owner, billing manager, app manager, security manager in public preview, custom roles); the current pages name no AI manager role. CLI OAuth includes authorization for organizations using SAML SSO. Without repository write access the cloud session is read-only; organization policies and SSO requirements apply. Repository read or write access sets session and automation rights. The requester cannot approve their own Copilot PR; personal vs shared automations differ in visibility, editing and identity. Seats are assigned; for card or PayPal customers seats are paid upfront and revoking a seat gives no prorated refund. Owners can suspend or uninstall apps and revoke seats. Content exclusion and organization policy management are only in Business and Enterprise.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Sourcedpartial
Podium defines member/admin instance roles and admin-only membership management. The released client authentication uses password-backed sessions and paired-device tokens without enterprise SSO.
Automation GitHub-tool writes can be scoped to all repository resources or the triggering issue. MCP tools keep separate credential scopes. Approval before tools that are not allowed; shell(COMMAND) allow and deny rules. Admin firewall allowlists; local per-host rules. The local proxy releases credentials only to approved HTTPS hosts. Workflows wait for Approve and run; humans review and merge draft PRs. Write-access users approve workflow runs; the requester cannot approve the PR. Allow once or for the session; allowlist changes apply to new sessions only. --allow-tool, --allow-all-tools and allow for the session. --deny-tool takes precedence; rejecting a request can include feedback. /sandbox disable and allowlist edits; revoking granted session approvals is not documented. The CLI asks by default; local sandboxing is off by default; the cloud firewall restricts internet access. Assisted permissions, plan mode and local sandboxing. Managed settings can require sandboxing and block when unavailable; enterprise agent-permission and MCP allowlists apply in the cloud.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Sourcedpartial
Agent-session management requests go through an operator approval broker with approve and deny; the privacy docs say lifecycle commands and automation requests may require operator approval. Harness permission prompts appear as a Permission needed state; credential copying during pairing is a per-grant opt-in. Command-level policy enforcement beyond the harness own modes was not established.
Shared automation definitions and runs are visible to repository readers; writers can edit them. Signed commits, session logs and audit log events. Admins see audit log events; approval records are not detailed. Issue artifacts (public preview) keep version history. Issue and PR comments. Memory and the session list persist across sessions. Admins export preferences; usage CSV; issue artifact download (public preview). 28 days for Actions sessions; account lifetime for cloud sandbox sessions. Commits are signed and Verified; the session-log integrity model is not documented.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Sourcedpartial
Issue actions, approval steps, comments and mail are recorded server-side. Issue activity events, including approval steps, are pruned by the server's janitor once they are older than 14 days or fall outside the newest 50,000 rows. That policy does not define all content retention, export or tamper-evident audit guarantees.
The app lets users review proposed session changes and PRs in-product. The cloud flow accepts refinements before the user opens a PR. My Work and PR filters in the app. App canvases show plans, terminals, browser previews and diffs. Copilot code review comments on PRs. Agent Merge fixes CI and conflicts and can pair with auto-merge. Max review effort (public preview) runs several reviewer agents plus an aggregator.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Review is a documented loop: issue stage review, diff and checks, artifacts, offers with merge or send-back buttons, and a Tray of review-ready work. Release 0.3.0 adds quoting text from a reply or review finding into the next message and a full-height reading view with an outline. The board has a key map with a help sheet. Guided-review quality and narrow-screen use were not measured.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
Shared automations belong to the repository and keep working after the creator leaves; a colleague continuing the same interactive session is not documented. Read-access users can view automation session prompts, logs and changes. Issue artifacts (public preview) stay on the issue. Cloud sandbox sessions continue without the owner connected. Shared automation actions are attributed to Copilot.
Checked 2026-11-21 (1 day old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
The durable issue, session transcript, artifacts and mail persist for any later agent, and a live session can be handed to another machine in an edge feature. Colleague takeover by a second person was not established.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
Project Library retains reference material and generated artifacts. Library collects a project’s artifacts. Project sessions link to changes, plans and associated PRs. The app Changes view and pull requests show diffs. Playwright takes screenshots; images attach to issue artifacts (public preview). gh issue artifact (issue artifacts, public preview) supports video attachments and automations can upload video. App canvases and Teams Canvas show previews. Security validation results appear in the session log. Copilot code review leaves comments and an overview comment. Each agent commit message links to its session log. Session data retention is 28 days or account lifetime by environment.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Artifact commands store immutable server-side copies of issue-linked files, including images, video, HTML and documents, which render in the issue sidebar and offers and stay readable after the worktree is gone. Each task has an artifacts page grouped into screens, documents and reports, with filters and search. Revision linking remains untested.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-07. Scope and sources
Cloud development sessions can run tests and linters; required delivery gates were not tested. Cloud agent runs tests and linters. Actions workflows run after approval; Agent Merge fixes failing workflows. Copilot code review gives a second opinion; Max effort (public preview) uses several providers. Branch protection and required checks apply to the agent. Rulesets and required checks block merge; the agent cannot merge its draft. Rulesets, Copilot approval settings and approvable paths. Copilot approval is dismissed on new commits; unattributed PRs need an extra approval.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
In the experimental Shipping path a configured validation profile runs before landing (default argv bun run test with a 10 minute timeout) and a failed validation holds the order. Lint, type, CI and spec conformance gates and independent review requirements were not found as product-enforced policy.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
CLI documentation gives commit and revert examples, separate from tested delivery behavior. Both Actions-based and cloud-sandbox sessions work on one branch and can open one PR per task. Further scoped facts are recorded in the atomic answers. Each task gets a copilot/ branch. Local app sessions run in their own worktrees. @copilot on a PR continues work on its branch. Agent Merge resolves merge conflicts. Sessions link to their PRs from the Agents tab.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Issues own branches and worktrees, and issue actions rebase, open a pull request with gh (pr create --fill) and merge, subject to a merge-style setting (ff-only by default, with auto-rebase before merge). Podium does not resolve branch conflicts, and PR review comments and CI response were not established.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
The app can leave a PR review; exact revision binding and enforced authority were not established. The app offers PR merge after reviewing changes and CI status.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-07. Scope and sources
Podium
Agent cross-checkedpartial
The operator accepts work through offers and runs the merge action under a branch merge-lock; in the experimental Shipping path approval is frozen to an exact head and delivery lands on a local target branch by default. Release and deployment are not product features found in the sources.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
Commit messages link to session logs. Sessions open PRs; Mobile can filter merged tasks. Mobile session filters include Open and Merged. Session logs and issue artifacts (public preview) are retained. New commits dismiss Copilot's approval.
Checked 2026-11-21 (1 day old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Shipping orders record approved base and head SHAs, tested integration SHA, landed ref and validation result in a durable delivery receipt, and stale approvals hold the order. Shipping is experimental. The pr action records a PR URL on the issue. Build and deployment links and rollback were not found.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
My Work and PR filters in the app. PR summaries on paid plans (not Free); review overview comment. App sessions are grouped by repository. PRs can be filtered by custom fields such as Size. Review findings stay on the PR. Issue artifacts (public preview) attach to each issue. Partner agents (Claude and Codex, public preview) iterate on PR comments. Copilot approval is dismissed when new commits are pushed, so it applies to the reviewed commits. Agent Merge, combinable with auto-merge. New commits dismiss approval.
Checked 2026-11-21 (1 day old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
The shipping service records revision-specific approval, configured validation and delivery receipts. The Shipping panel is an experimental edge toggle, off by default; the separate Queues panel was removed.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-07. Scope and sources
Copilot application/service use is governed by custom GitHub or Microsoft terms according to purchaser and plan. Hosted Copilot service use is covered by the plan-specific service terms, not a public SDK license. The Copilot SDK (microsoft/copilot-sdk) and GitHub Agentic Workflows (github/gh-aw) are MIT licensed; the Copilot service, CLI and app are not open source. The github/app and github/copilot-cli repositories host releases, issues and licence files only, not source; the VS Code Copilot extension source is public (MIT) in microsoft/vscode/extensions/copilot. At this check: app v1.4.3 (2026-11-19; v1.2.0 released 2026-11-10), CLI v1.0.114 stable (2026-11-18) and v1.0.115-5 prerelease, SDK v2.0.1 stable (2026-11-18) and v2.0.2-preview.4 prerelease (microsoft/copilot-sdk), VS Code Copilot extension source in microsoft/vscode at bfbf12b.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
The public repository is licensed Apache 2.0, and the website presents Podium as an open-source agent development environment. Hosted extensions and hosted-service terms are not established.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
Copilot CLI can run on user infrastructure without a GitHub account in offline BYOK mode, using a local/in-network model. Hosted Copilot UI and integrations are separate. Offline mode makes no GitHub requests; a remote model endpoint still receives context if configured. BYOK CLI does not require GitHub login. Admins export user preference memories; issue artifacts (public preview) can be downloaded with gh issue artifact; AI usage is exported as CSV reports. Copilot CLI bulk-imports prepared memories. Issue artifacts (public preview) are Markdown or HTML files; creating and viewing HTML artifacts in the GitHub app is coming soon. Usage reports are CSV. Individual plan cancellation takes effect at the end of the billing cycle; usage incurred before cancelling is still owed.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Local desktop and self-hosted installs run on user-controlled machines with a documented state directory. The docs describe stopped full-state backup and automatic pre-migration database-only snapshots retaining three copies. Those snapshots omit artifacts, transcripts, uploads, configuration and secrets; no portable structured export/import contract was established.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-11. Scope and sources
COPILOT_OFFLINE=true fully disables CLI telemetry and GitHub network requests; configured model-provider traffic remains. Offline BYOK can use local or in-network inference; an internet provider still receives prompts and code. Cloud sessions clone repositories into GitHub-hosted sandboxes. Chat and session data in cloud sandbox sessions is kept for the life of the account; Actions-based sessions keep data 28 days. Local sandboxing gives tools credential placeholders and a local proxy supplies real credentials only to approved hosts; personal MCP secrets are stored in Copilot settings. Session logs show the agent's work and tools; session logs and audit log events are available to administrators. Issue artifacts are stored on the issue (public preview). Session data: 28 days (Actions) or account lifetime (cloud sandbox); memories expire after 90 days by default. Memories can be deleted by users and administrators; session deletion was not established in these sources. The app signs in to GHE.com data-residency hosts; the cloud sandbox region is not stated. The OpenTelemetry identity attributes and five capture classes are documented. OpenTelemetry identity capture and all content capture classes are off by default; a separate pseudonymous user identifier exists.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Docs state repositories and native harness credentials stay on machines you control and server state lives in a documented directory. Since 0.2.0 a daily usage report is on by default and can be turned off; crash reports stay opt-in. Usage reports carry an install id, schema version, Podium version, OS, architecture, bucketed install age and machine count, session counts per harness and whether issues were used. Crash reports carry the same install id, versions, OS and architecture, a fixed error type and up to 20 stack frames, each with a Podium source file path, line and function when there is one; error messages and frames outside Podium are dropped. Neither tier sends prompts, transcripts, your code or paths outside the Podium source tree. Model-provider handling is the harness own. Hosted analytics and website analytics are separate paths.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-21. Scope and sources
At the 7 November check: Pro $10/month, Pro+ $39/month, Max $100/month, Business $19/granted seat/month and Enterprise $39/granted seat/month, in USD. Pro lists 1,000 base plus 500 flex credits; Pro+ 3,900 plus 3,100; Max 10,000 plus 10,000. Credit amounts are not task counts. Cloud-sandbox compute, memory and snapshot storage are billed by consumption after the shared allowance. The CLI can use a user-provided model provider API key, paid separately to that provider. Copilot Enterprise is described as designed for GitHub Enterprise Cloud enterprises, and Copilot access is $0 for GitHub Enterprise Cloud licensees; the cost of a required GitHub Enterprise Cloud plan is not stated on these pages. Usage beyond the included credits costs $0.01 per AI credit through an additional-usage budget.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
No Podium price is published: podium.do/pricing returns 404 and the docs list billing as not implemented. The software is open source and free to download; model and compute costs are the user's own.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
One person: Pro $10, Pro+ $39 or Max $100 per month, plus credits beyond the allowance at $0.01 and cloud sandbox compute. Billing for cloud sandbox usage begins 2026-11-23. Business for two: 2 x $19 = $38 per month with 3,800 pooled credits. Business for five: 5 x $19 = $95 per month with 5 x 1,900 = 9,500 pooled credits; overage $0.01 per credit. Token rates are published per 1M tokens per model, for example Claude Sonnet 5.5 $2 input and $10 output. Cloud sandbox compute is $0.000024 per vCPU second ($0.0864 per vCPU-hour) and memory $0.000003 per GiB second ($0.0108 per GiB-hour); sandbox sizes are not published. Billing for cloud sandbox usage begins 2026-11-23. Published seat prices: Pro $10, Pro+ $39, Max $100 per month; Business $19 and Enterprise $39 per granted seat per month. Variable costs are AI credits beyond the allowance and cloud sandbox compute, memory and storage. Billing for cloud sandbox usage begins 2026-11-23. Plans include AI credit allowances and cloud sandbox use draws from the included compute allowance. Billing for cloud sandbox usage begins 2026-11-23. Task token counts, sandbox vCPU and memory sizes and the Free credit amount are not published.
Checked 2026-11-21 (1 day old); agent cross-checked 2026-11-21. Scope and sources
Podium
No sourced public findingnot-published
No publishable sourced finding is available for this criterion. This does not mean the product lacks the capability.
Copilot is not available for GitHub Enterprise Server; model and feature availability depend on policies and spend controls. Paid plans and Copilot Business/Enterprise contact GitHub Support; GitHub Free uses Community discussions; preview features are out of support scope. GitHub Status lists Copilot and Copilot AI Model Providers components; incidents include critical 2026-10-28 and 2026-11-11 and major 2026-11-09. Models were retired on 2026-10-19 and 2026-11-16, with further deprecations announced for December; the SDK is moving to Microsoft as Microsoft Copilot SDK. Exports found: issue artifacts (public preview), administrator memory export and usage reports.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Podium is pre-1.0. Stable v0.3.1 (2026-11-06), an update fix, follows v0.3.0 (2026-11-05), v0.2.0 (2026-10-11) and v0.1.0 (2026-08-21), with edge and dev prerelease channels. Windows and Linux desktop are Experimental, the native iPhone app is in private beta and hosted Podium is unconfirmed. Support runs through GitHub issues, Discord and private vulnerability reporting.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
A SOC 2 Type 2 report is listed for enterprises and the 2024 post confirms Type II Copilot coverage; newer Copilot periods were not verified. Copilot Business and Enterprise were in scope of the 2024 SOC reports. 2024-04-01 to 2024-09-30 is the latest period stated in a public first-party post; later periods are not publicly stated. Enterprise owners download reports from the enterprise Compliance page; the 2024 post points to the GitHub Enterprise Trust Center. Public first-party statements checked 2026-11-21; reports themselves not inspected. Evidence is vendor statements; the gated reports were not inspected.
Checked 2026-11-21 (1 day old); agent cross-checked 2026-11-21. Scope and sources
Podium
No sourced public findingnot-published
No publishable sourced finding is available for this criterion. This does not mean the product lacks the capability.
Free has limited features and an AI-credit allowance; the plans page says all plans include CLI and the Copilot app. Copilot in the cloud requires a paid plan; Free retains GitHub.com and GitHub Mobile chat. BYOK and local models work without a Copilot plan; provider costs apply. Projects, partner agents and the tasks API are public preview on eligible plans. Free covers the app and CLI (including with BYOK) but not Copilot in the cloud; automations require Pro or higher.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
Agent cross-checkedpartial
Podium is open-source Apache-2.0 software, and the download page calls it free and open source. Download links for macOS, Linux and Windows redirect to GitHub release assets without a login (observed by HTTP requests; not installed). No Podium charge is documented for self-hosting; agent provider usage is separate. The native iPhone app is private beta and hosted access is unconfirmed.
Checked 2026-11-06 (16 days old); agent cross-checked 2026-11-06. Scope and sources
Business/Enterprise include pooled credits with separately billed excess usage; Copilot access has no included usage allowance. Overage is $0.01 per AI credit. Partially specified: base credits, overage and model token rates are published, but the Free credit amount, flex allotment stability, sandbox sizes and code review model are not.
Checked 2026-11-07 (15 days old); agent cross-checked 2026-11-21. Scope and sources
Podium
No sourced public findingnot-published
No publishable sourced finding is available for this criterion. This does not mean the product lacks the capability.