canon

Use when you need a governed Canon review of a real diff or pull-request range instead of a loose chat summary.

Use when a repository does not have Canon runtime state yet and you need to initialize .canon before any governed workflow.

Use when you need a governed Canon policy-shaping run to shape a new or modified policy with mandatory impact evaluation.

Use when you need a governed backlog run that decomposes bounded upstream decisions into delivery epics and slices.

Use when you need a governed Canon architecture run to record decisions, tradeoffs, and risk-gated approvals.

Use when you need a governed change run in a live codebase where invariants and existing behavior matter.

Use when you need a governed domain-language packet that stabilizes the shared vocabulary of a product area before downstream design or change work.

Use when you need a governed domain-model packet that formalizes domain concepts, relationships, invariants, and feature-impact rules before architecture or backlog decomposition.

Use when you need a governed migration packet for an existing system with explicit compatibility, sequencing, and fallback posture.

Use when a Canon run is complete and the user wants to publish the packet into docs or specs from chat.

Create or update the project constitution from interactive or provided principle inputs, ensuring all dependent templates stay in sync.

Use when you need a governed change run in a live codebase where invariants and existing behavior matter.

Use when a Canon run is complete and the user wants to publish the packet into docs or specs from chat.

Use when you need a governed Canon discovery run to bound a problem space before requirements or delivery planning.

Use when you need a governed implementation run for an existing system with explicit task mapping and mutation bounds.

Use when you need a governed incident packet for an existing system with explicit blast-radius, containment, and follow-up readiness.

Use when you need a governed refactor run for an existing system with preserved behavior and explicit no-feature-addition evidence.

Use when you need a bounded requirements run in Canon before code, architecture, or execution drift starts.

Use when a real Canon run has been unblocked and you need to continue it from recorded runtime state.

Use when you need a governed Canon review of a bounded non-PR change package or artifact bundle.

Use when a Canon run already exists and you need its current state, pending approvals, and next steps.

Use when you need a governed Canon system-shaping run to shape a new system with mandatory critique and persisted evidence.

Use when you need a governed Canon verification run to challenge claims, invariants, contracts, or evidence directly.

Use when you need a governed implementation run for an existing system with explicit task mapping and mutation bounds.

Use when you need Canon-backed missing-context findings, explicit output-quality posture, materially-closed decision signals, and targeted clarification questions from authored file-backed mode inputs before starting a run.

Use when you need a governed system-assessment packet for an existing system with ISO 42010 coverage, explicit observed findings, inferred findings, and assessment gaps, and publishable evidence.

Use when you need a governed supply-chain-analysis packet for an existing repository with explicit SBOM, vulnerability, license, and legacy posture evidence.

Use when you need a governed security-assessment packet for an existing system with explicit threats, risks, mitigations, and evidence gaps.

Use when Canon has blocked or gated a real run and you need to record an explicit approval against the actual runtime target.

Use when a Canon run already exists and you need the emitted artifact paths rather than run state or evidence lineage.

Use when you need the evidence bundle, lineage, and linked runtime surfaces for an existing Canon run.

Use when you need request-level Canon decisions, attempts, and policy outcomes for an existing run.

Create or update the feature specification from a natural language feature description.

Generate an actionable, dependency-ordered tasks.md for the feature based on available design artifacts.

Execute the implementation planning workflow using the plan template to generate design artifacts.

Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.

Execute the implementation plan by processing and executing all tasks defined in tasks.md