Entity identities, frontmatter, host availability, and exact source embeds regenerate from the shipped payload on every build. Curated operating guidance is hand-reviewed and contract-tested, but it can still lag a source change; when the two disagree, the exact source embed is authoritative. See how the three catalogs cooperate in How a Request Flows: a command routes to an owning skill, which may dispatch specialist agents.
Use the catalog from left to right.
- Choose the public command that matches the outcome you need.
- Follow its owning skill to understand phases, stops, and durable artifacts.
- Open an agent only to inspect a dispatched role's tools and constraints; agents are not a second command surface.
Every entity page begins with host-native syntax or dispatch context, then curated operating guidance, gates, related routes, and the exact source used to generate it.
Commands, skills, and agents below are generated from the ca (Claude Code) plugin payload. Each command page checks the shipped Claude Code, Codex, and Pi catalogs and marks adapters where that command is not shipped. The Agents catalog below describes Claude Code’s packaged agent charters. Current Codex releases load those charters into host-provided agent threads and retain thread receipts; older hosts may fall back to inline execution except where isolated scouts are mandatory. Pi dispatches hardened children through the parent-only codearbiter_dispatch tool. See Compatibility → Host Differences for the full per-surface breakdown across all three hosts.
| Command | Description |
|---|
| preview | Zero-onboarding, read-only dry-run of the reviewer fleet against the current uncommitted diff. Predicts reviewers, runs the state-free secret scan, writes nothing. |
| Command | Description |
|---|
| init | Opt this repo into codeArbiter — scaffold the root-level .codearbiter/ state store. |
| Command | Description |
|---|
| add-dep | Vet a new or changed third-party dependency for license, provenance, and supply-chain risk before any install runs. |
| chore | Sanctioned lane for non-behavioral work — docs-only edits, dependency bumps, reverts. Type-scaled gates; no TDD demanded of prose. |
| feature | Start a feature: brainstorm a spec, get it approved, then drive it test-first through the pipeline. The one entry to implementation. |
| fix | Fix a confirmed bug: a failing regression test first, then a minimal fix, then the rest of the tdd gates. |
| refactor | Restructure code with behavioral parity proven through unmodified pre-existing tests, then refactor. No behavior change. |
| spike | Exploratory spike on a throwaway branch — answer a named question with disposable code. Never merges; exits to a findings note or /ca:feature. |
| sprint (preview) | Autonomous sprint — one interactive spec gate, then plan-to-PR execution with every auto-decision SMARTS-scored and logged. Hard gates remain true stops. |
| Command | Description |
|---|
| review | Review a diff with the reviewer fleet, funneled to one triaged verdict. Targets the current working diff, a path, or an inbound GitHub PR. |
| Command | Description |
|---|
| adr | Author a numbered, dated, user-attributed Architecture Decision Record under .codearbiter/decisions/. |
| Command | Description |
|---|
| commit | Run the full commit gate — the only sanctioned path to a git commit. |
| pr | Open a pull request the only sanctioned way — clear every BLOCK-level review finding, then stage the PR. Never a direct write to the default branch. |
| release | Cut a release the only sanctioned way — derive the target’s declared version policy from the commit log, roll its changelog, compose an annotated tag, and optionally publish its exact declared assets. Takes the declared target’s name as its only argument, or —dry-run to preview one with no write. The only path to a version tag. |
| Command | Description |
|---|
| doctor | Verify the active host install, package, command ownership, enforcement, and harmless live-fire probe. Read-only. |
| override | Sanctioned, logged bypass of a gate or hard rule — one audit line, then proceed. |
| status | Show the project’s current state at a glance — stage, open tasks, open questions, overrides since the last checkpoint, current branch. Read-only. |
| task | The sanctioned task-board mutator — add a queued task, start one (flips to in-progress and stamps the date, minting a dotted ID on pick-up), or mark an in-progress task done. The only blessed write to open-tasks.md. |
| Command | Description |
|---|
| debug | Investigate-then-decide root-cause analysis for a defect whose cause is unknown. No code changes — exits to /ca:fix, /ca:adr, or a no-action close. |
| Command | Description |
|---|
| checkpoint | Periodic multi-reviewer sweep of the whole codebase — surfaces a triaged checkpoint report. |
| threat-model | Opt-in lightweight STRIDE pass for a sensitive feature before implementation. Not a routine gate — invoke it when a change warrants security thought. |
| tribunal | Deep, rarely-convened whole-codebase audit — eleven specialist lenses, a resumable on-disk audit log, findings filed as GitHub issues on approval. Expensive; estimates cost and STOPs before running. Never a required gate. |
| Command | Description |
|---|
| adr-status | Report the health of Architecture Decision Records — aged, unchallenged, supersession candidates, unresolved CONFIRM-NN. Read-only. |
| reconcile | SMARTS arbitration — reconcile architectural artifacts against the scaffold and prior decisions; every variance resolved by an explicit, user-attributed choice. |
| Command | Description |
|---|
| audit | Assemble the governance record for a range — commits, overrides, ADRs, sprint auto-decisions, open questions, checkpoint findings — into one dated audit packet. Read-only. |
| metrics | Read-only 3-metric governance glance — override rate, small-lane rate, sprint low-confidence ratio — each with a trend arrow vs. the prior 20-commit window. |
| prune (preview) | Trim transcript clutter to extend session lifetime — analyze, prune a copy, or toggle the after-each-turn service. Dry-run by default; gains land at resume/compaction, not the current turn. |
| standup | Daily repo hygiene — review the day’s repo state, then perform the cleanups under per-action confirmation. Fast-forward only, never destructive without a yes. |
| statusline | Wire codeArbiter’s statusline into ~/.claude/settings.json, or remove it. |
| Command | Description |
|---|
| new-skill | Author a new codeArbiter skill: prove the gap is real, get the spec approved, then write it. |
| Command | Description |
|---|
| commands | Show the codeArbiter command catalog — the public command list and what each routes to. |
| Command | Description |
|---|
| create-context | Brownfield back-fill — scout an existing codebase and populate .codearbiter/, then lock it initialized. |
| decompose | Greenfield decomposition interview — a layered interview that populates .codearbiter/ and locks it initialized. |
| Command | Description |
|---|
| cleanup | Finish an already-merged branch — classify the leftover artifacts, return to a fast-forwarded default checkout, and delete the merged local branch. Every discard confirmed per item; containment proven, never assumed. |
| watch | Watch a PR’s CI to completion — diagnose on red, notify and offer the merge on green. Never auto-merges. |
| Command | Description |
|---|
| context-check | Optional manual drift audit — report stale provenance-tracked docs, then per stale doc offer re-scout, re-baseline, or defer. Not the daily loop; commit-gate auto-heal owns routine maintenance. |
| Command | Description |
|---|
| conflict | Stop everything and surface a rule conflict — persona vs. docs vs. code. Present both sides and the conflict-hierarchy level; the user resolves. No silent reconciliation. |
| Command | Description |
|---|
| btw | Lightweight Q&A about the project — answer from context and return, no routing, no state change. |
| Skill | Description |
|---|
| brainstorming | The Socratic spec-refinement front of /feature, and the planning front of /sprint. |
| commit-gate | The only path to a commit. |
| context-check | Optional manual drift audit — report stale provenance-tracked docs (via _provenancelib drift detection across .codearbiter/.provenance/), then per stale doc offer re-scout / re-baseline / defer. |
| context-creation | The brownfield back-fill. |
| crypto-compliance | The banned-primitive gate. |
| debug | Investigate-then-decide root-cause analysis for a defect whose cause is unknown (distinct from /fix, which assumes a known bug). |
| decision-lifecycle | Author and track Architecture Decision Records. |
| decision-variance | Reconcile the project’s architectural artifacts against the scaffold and prior decisions, then present each variance as a SMARTS analysis for the user to decide. |
| decompose | The greenfield decomposition interview. |
| dispatching-parallel-agents | The parallel fan-out primitive. |
| executing-plans | The checkpoint coordinator for /feature. |
| finishing-a-development-branch | The terminal step of /feature and /sprint. |
| post-merge-cleanup | Finish an already-merged branch. |
| refactor | The behavior-preserving restructure gate. |
| release | The single permitted path to a version tag, for any target declared in the project’s .codearbiter/release-targets.md. |
| secret-handling | The secret-source gate. |
| security-architecture | Optional, opt-in STRIDE threat pass for a sensitive feature — invoked deliberately via /threat-model, never forced on ordinary changes. |
| skill-author | The authoring gate for new skills. |
| subagent-driven-development | The implementation engine. |
| tdd | The test-first gate. |
| tribunal | The deep, rarely-convened whole-codebase audit lane. |
| using-git-worktrees | OPTIONAL per-task isolation for autonomous parallel work. |
| writing-plans | The spec-to-plan bridge. |
| Agent | Model tier | Description |
|---|
| architecture-drift-reviewer | Haiku | Read-only checkpoint reviewer. |
| auth-crypto-reviewer | inherit | Reviews authentication, cryptography, key handling, and secrets against the repository’s .codearbiter/security-controls.md. |
| backend-author | Sonnet | Use when writing or modifying backend/server-side code. |
| checkpoint-aggregator | Haiku | Persists a complete verdict-aggregator output under the selected dated checkpoint filename in .codearbiter/checkpoints/. |
| coverage-auditor | Haiku | Dispatched by the tdd skill (Phase 4) to audit test coverage against TDD obligations. |
| decision-challenger | inherit | Adversarial red-team reviewer of ADRs. |
| dependency-reviewer | Sonnet | Dispatched when package.json, lock files, or container base images change. |
| design-quality-reviewer | Sonnet | Reviews generated, user-facing visual or formatted output (UI, reports, slides, charts, diagrams, CLI output) against the anti-slop-design reference. |
| finding-triage | Haiku | Post-processes reviewer reports — consolidates findings, classifies each by severity and whether it blocks the current change. |
| frontend-author | Sonnet | Use when writing or modifying frontend/UI code. |
| grader | inherit | INTERNAL SMARTS analyst dispatched by the decision-variance skill. |
| infra-author | Sonnet | Use when writing or modifying IaC, containers, CI/CD manifests, or deployment configuration. |
| map-deps | inherit | Dispatched by the tribunal deep-audit lane’s Phase 1, on a large/sprawling repo, to offload dependency/integration-surface mapping out of the orchestrator’s retained context. |
| map-structure | inherit | Dispatched by the tribunal deep-audit lane’s Phase 1, on a large/sprawling repo, to offload structural mapping out of the orchestrator’s retained context. |
| migration-reviewer | inherit | Dispatched when a database migration file is added or modified. |
| scout | Haiku | INTERNAL evidence-gatherer dispatched by the decision-variance and context-creation skills. |
| security-reviewer | inherit | Dispatch PROACTIVELY when a change touches authentication, authorization, cryptography, secrets, deployment manifests, network policies, or CI workflows. |
| tribunal-lens-reviewer | inherit | Dispatched by the tribunal deep-audit lane, once per active lens, with a lens assignment. |
| verdict-aggregator | Haiku | Composes a complete finding-triage report into one structured in-memory verdict for the caller. |
Each lens card is the per-lens mandate the tribunal-lens-reviewer agent executes when /ca:tribunal dispatches it under that lens’s assignment.
| Lens | Description |
|---|
| appsec | The appsec tribunal lens: the per-lens mandate the tribunal-lens-reviewer agent executes when a /ca:tribunal deep audit activates this lens. |
| architecture | The architecture tribunal lens: the per-lens mandate the tribunal-lens-reviewer agent executes when a /ca:tribunal deep audit activates this lens. |
| coverage | The coverage tribunal lens: the per-lens mandate the tribunal-lens-reviewer agent executes when a /ca:tribunal deep audit activates this lens. |
| infra | The infra tribunal lens: the per-lens mandate the tribunal-lens-reviewer agent executes when a /ca:tribunal deep audit activates this lens. |
| migration | The migration tribunal lens: the per-lens mandate the tribunal-lens-reviewer agent executes when a /ca:tribunal deep audit activates this lens. |
| observability | The observability tribunal lens: the per-lens mandate the tribunal-lens-reviewer agent executes when a /ca:tribunal deep audit activates this lens. |
| performance | The performance tribunal lens: the per-lens mandate the tribunal-lens-reviewer agent executes when a /ca:tribunal deep audit activates this lens. |
| reliability | The reliability tribunal lens: the per-lens mandate the tribunal-lens-reviewer agent executes when a /ca:tribunal deep audit activates this lens. |
| secrets-supply | The secrets-supply tribunal lens: the per-lens mandate the tribunal-lens-reviewer agent executes when a /ca:tribunal deep audit activates this lens. |
| test-fidelity | The test-fidelity tribunal lens: the per-lens mandate the tribunal-lens-reviewer agent executes when a /ca:tribunal deep audit activates this lens. |
| typesafety | The typesafety tribunal lens: the per-lens mandate the tribunal-lens-reviewer agent executes when a /ca:tribunal deep audit activates this lens. |
