🦀

cargo install decapod && decapod init

Decapod
A daemonless control plane for AI coding agents.

Called on demand inside agent loops. No background process, no new workflow, local-first state you can verify.

---

Why Decapod 🧠

AI coding agents are strong at generating code. Most failures happen before and after generation: unclear intent, fuzzy boundaries, and weak completion checks.

Decapod is the missing layer in that loop. Agents call it mid-run to lock intent, enforce boundaries, and prove completion with explicit gates. It shapes inference without doing inference.

Decapod is daemonless. There is no long-lived service. The binary starts when an agent calls it and exits immediately after the call.

"Just use Decapod" is literal:

• cargo install decapod
• decapod init

Then continue with Claude Code, OpenAI Codex, Gemini CLI, Cursor, or any tool that can invoke a CLI command. Decapod is agent-agnostic and safe for concurrent multi-agent execution.

State is local and durable in .decapod/: shared context, decisions, and traces persist across sessions and remain retrievable over time.

Related: Evaluating AGENTS.md (ETH SRI, 2026) on context-file quality and agent cost/performance.

☕ Like Decapod? Buy us a coffee on Ko-fi 💙

Assurance Model ✅

Decapod centers execution around three outcomes:

• Advisory: clear next actions that tighten intent and reduce wasted loops.
• Interlock: hard policy boundaries that block unsafe or out-of-contract flow.
• Attestation: durable, structured proof that completion criteria actually passed.

Operating Model ⚙️

Human Intent
    |
    v
AI Agent(s)  <---->  Decapod Runtime  <---->  Repository + Policy
                         |    |    |
                         |    |    +-- Interlock (enforced boundaries)
                         |    +------- Advisory (guided execution)
                         +------------ Attestation (verifiable outcomes)

Features ✨

• ✅ Daemonless by design: no background service, no long-lived process, no hidden control loop.
• ✅ cargo install decapod + decapod init, then keep using your agents exactly as-is.
• ✅ Scoped plain-English policy overrides in .decapod/OVERRIDE.md.
• ✅ Agent-agnostic control plane: one CLI/RPC contract across Claude, Codex, Gemini, Cursor, and more.
• ✅ Multi-agent-safe in one repo with governed task and workspace coordination.
• ✅ Shared transient skills memory: capture human-taught preferences once and apply them across agents/sessions.
• ✅ Work Unit manifests make intent/spec/state/proof explicit and durable.
• ✅ VERIFIED is structural, not rhetorical: proof-plan results must pass before completion.
• ✅ Variance-aware eval kernel (eval plan/ingest-run/judge/aggregate/gate) adds repeatable settings, judge contracts, and statistical regression gates.
• ✅ Deterministic context capsules (core|interfaces|plugins) return scoped, hashable guidance.
• ✅ Knowledge promotion firewall ledger (.decapod/data/knowledge.promotions.jsonl) controls what can become procedural truth.
• ✅ Procedural knowledge writes require event-backed provenance, blocking unproven memory drift.
• ✅ Local-first .decapod/ state keeps shared traces, decisions, and proofs durable and auditable.

And dozens more. For the full high-level and data-level surface area, see decapod docs show core/INTERFACES.md and the override template at .decapod/OVERRIDE.md.

Getting Started 🚀

cargo install decapod
decapod init

Then keep using your agents normally. Decapod is called from inside those agent runs when control-plane decisions are needed.

Agent integration: If you use Claude Code / Codex / Gemini / Cursor / similar tools, see AGENTS.md and the tool-specific entrypoint files (CLAUDE.md, CODEX.md, GEMINI.md) for the exact operational contract.

Learn more about the embedded constitution via the CLI:

decapod docs show core/DECAPOD.md

Override constitution defaults with plain English in .decapod/OVERRIDE.md.

Contributing 🤝

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

Documentation 📚

• Development guide: CONTRIBUTING.md
• Security policy: SECURITY.md
• Release history: CHANGELOG.md

Support 💖

• 🐛 File an issue
• ☕ Support on Ko-fi

License 📄

MIT. See LICENSE.