This is a Decapod-managed repository. Strict Dependency: You are strictly bound to the Decapod governance kernel.
Before performing any Decapod operations beyond help/version checks, you MUST ingest the Agent API Orientation corpus from the Decapod binary:
decapod docs ingestThis command outputs the embedded agent documentation (api-index.md, command-contracts.md, payload-examples.md, error-recovery.md, state-model.md, config-schema.md, mcp.md) directly from the binary — no filesystem reads required.
Note on Constitution vs Docs: Agents use decapod constitution to discover technology, primitives in software engineering and delivery, and for scoping pre-inference context. Agents use decapod docs to access the embedded agent-facing methodology documentation (restricted to docs/agent/).
# Smart bootstrap: only install/init if a newer version is available
(decapod capabilities --format json | grep -q '"is_latest":true') || (cargo install decapod && decapod init --proof)
decapod docs ingest && decapod validate && decapod session acquire
decapod rpc --op agent.init
decapod workspace status
decapod todo add "<task>" && decapod todo claim --id <task-id>
decapod infer orientation --task-id <task-id>
decapod workspace ensure
cd .decapod/workspaces/<your-worktree>
decapod constitution get core/DECAPOD
decapod rpc --op context.resolvedecapod capabilities --format json
decapod data schema --deterministic
decapod infer orientation --intent "<your-goal>" --task-id <id>
decapod govern capsule query --topic "<topic>" --scope interfaces --task-id <task-id>
decapod constitution search --query "<problem>"Decapod is pre-inference faculty, not post-hoc validation.
- You MUST consult Decapod early in the conversation, before hardening a plan, selecting an architecture, choosing tools, writing implementation code, or claiming confidence about ambiguous user intent.
- Use Decapod when the user asks for an app, feature, service, workflow, architecture, data model, integration, security change, deployment path, API, UI, or production-quality implementation.
- Start with
core/DECAPODfor broad prompts. Usecore/*nodes as secondary routers. Use non-core nodes as institutional doctrine. - After retrieval, choose one of three states: ask the user a sharper question, query Decapod again, or proceed with explicit assumptions and proof expectations.
- Do not wait until after code is written to discover that the work violated intent, boundaries, proof, or institutional standards.
- MUST refine intent with the user before inference-heavy work.
- MUST use
decapod infer orientationbefore non-trivial implementation. - MUST stop and ask the human when Decapod emits a Decision Gate.
- MUST create and claim a Decapod todo before
decapod workspace ensure,decapod workspace ensure --container, or any container run. - MUST NOT work on main/master or modify the root repository's active branch. MUST use
decapod workspace ensure. - MUST read .decapod/config.toml as user-editable project context.
- MUST NOT claim done without
decapod validatepassing. - MUST NOT invent capabilities that are not exposed by the binary.
- MUST stop if requirements conflict or intent is ambiguous.
- MUST respect the interface abstraction boundary.
- MUST maintain Living Specs: treat
.decapod/generated/specs/*as dynamic documents. - MUST use the command contracts from
decapod docsoutput instead of guessing arguments.
Agents act. Decapod orients. Call Decapod at decision boundaries: ambiguous requests, public impact, unclear proof, todo lifecycle, scope expansion, context loss, or multi-agent collision risk.
The files under .decapod/generated/specs/ are living contracts. Review and update INTENT.md, ARCHITECTURE.md, and INTERFACES.md to align with evolving intent and reality.
Preserve the chain between intent, context, assumptions, action, and proof.
- Preserve Uncertainty: Summaries must preserve risk instead of compressing it.
- Recursive Continuity: Prior assumptions MUST carry forward until resolved.
- Evidence-Based Claims: Claims of completion must be tied to measured evidence.
- Clarification Trigger: Stop if a critical assumption cannot be proven.
- INV-DAEMONLESS: Decapod MUST NOT leave background processes running.
- INV-BOUNDED-VALIDATE:
decapod validateMUST terminate within bounded time. - INV-STORE-BOUNDARY: Agents MUST NOT directly mutate
.decapod/*. - INV-SESSION-AUTH: Mutations require active session.
- INV-PROOF-GATED:
VERIFIEDstatus requires passed proof-plan gates. - INV-ROOT-ISOLATION: Agents MUST NOT mutate files in the main repository checkout.
- ✅ Router:
decapod constitution get core/DECAPOD - ✅ Gates:
decapod capabilities|decapod docs ingest|decapod validate - ✅ Claims:
decapod todo claim --id <task-id>| ✅ Auth:DECAPOD_SESSION_PASSWORD - ✅ Workspace: Docker git workspaces | ✅ Privilege: request elevated permissions before Docker/container workspace commands
- Read
.decapod/config.tomland.decapod/OVERRIDE.md. - External task managers are allowed, but Decapod todos are still mandatory for workspace isolation.
- DO NOT mutate
.decapod/state directly; all access MUST use CLI. Access to.decapod/is strictly via decapod CLI. - Stop if requirements conflict or intent is ambiguous. Respect the interface abstraction boundary.