decapod 0.5.2

Decapod is a Rust-built governance runtime for AI agents: repo-native state, enforced workflow, proof gates, safe coordination.
Documentation

Demo: Watch on GitHub (12 MB)


What Is Decapod

Like Docker is a runtime for containers, Decapod is a runtime for agents. You set it up once, then agents operate inside a governed environment with persistent state, enforced methodology, proof gates, and coordination primitives.

Agents can write code. But they can't reliably ship because they forget what they built yesterday, treat best practices as vibes, say "done" without evidence, and trip over each other in parallel. Decapod fixes that.

Decapod is not a prompt pack, an agent framework, a hosted SaaS platform, or a review bot. It's infrastructure: the environment where agent work becomes enforceable.

Quickstart

cargo install decapod
cd your-project
decapod init

That's it. Agents now operate inside the governed environment. You observe outcomes, review summaries, and merge when proofs pass.

How It Works

Persistent state — Agents persist work to .decapod/: todos, conventions, decisions, proof events. Durable state that survives sessions and model switches.

Enforced methodology — An embedded constitution defines binding contracts for how agents operate. Generated entrypoints (CLAUDE.md, AGENTS.md, GEMINI.md, CODEX.md, OPENCODE.md) require agents to read the constitution, use the control surface, and follow Intent > Architecture > Implementation > Proof. Projects customize via .decapod/OVERRIDE.md.

Proof gates — Agents must satisfy validation gates before claiming completion. If decapod validate fails, the work isn't done. Evidence required, not assertions.

Coordination — A shared backlog with audit trail, shared conventions and preferences, a proof ledger, and policy boundaries. Multiple agents work in parallel without collisions.

CLI Reference

decapod init        Bootstrap a project                    (alias: i)
decapod docs        Access methodology documentation       (alias: d)
decapod todo        Track tasks and work items              (alias: t)
decapod validate    Validate methodology compliance         (alias: v)
decapod govern      Governance: policy, health, proofs      (alias: g)
decapod data        Data: archives, knowledge, context
decapod auto        Automation: scheduled, event-driven     (alias: a)
decapod qa          Quality assurance: verification         (alias: q)

Architecture

your-project/
├── AGENTS.md              Agent universal contract (generated)
├── CLAUDE.md              Claude entrypoint (generated)
├── GEMINI.md              Gemini entrypoint (generated)
├── CODEX.md               Codex entrypoint (generated)
├── OPENCODE.md            OpenCode entrypoint (generated)
└── .decapod/
    ├── data/              State: SQLite DBs + event logs
    ├── generated/         Derived artifacts (auto-managed)
    ├── OVERRIDE.md        Project-specific constitution overrides
    └── README.md          Control plane documentation

Agents interact through the CLI control surface. You don't touch .decapod/data/ directly.

Security

See SECURITY.md. Agents must handle credentials securely — never log, never commit, always rotate. Violations are constitutional breaches.

Contributing

Decapod is built with Decapod. To develop:

git clone https://github.com/DecapodLabs/decapod
cd decapod
cargo build
cargo test
decapod validate

Support

License

MIT. See LICENSE for details.