govctl CLI
A CLI that keeps RFCs, ADRs and work items as TOML in your repo and stops an agent from closing work until checks pass
- 276 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Edits files in the gov/ folder and runs the checks defined in project configuration
- The integration adds skills, hooks and reviewers to Claude Code and Codex
- The self-update command replaces its own binary
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add cli/govctlDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Edits files in the gov/ folder and runs the checks defined in project configuration
- The integration adds skills, hooks and reviewers to Claude Code and Codex
- The self-update command replaces its own binary
This is third-party code. Review the repository files before installing.
What it does
govctl is a command-line program for teams that build software with AI agents and want a document behind every change. In the repository's `gov/` folder it keeps RFCs (what must be true), ADRs (why a design was chosen), work items with acceptance criteria, verification guards and conformance cases that tie scenarios to versioned requirements. An agent works through stable commands: `govctl rfc get`, `govctl adr edit`, `govctl work move` and others. Artifacts are TOML files, so changes show up in diffs and reviews. A work item closes when its configured guards pass, and the agent saying it is done decides nothing. `govctl search` looks across all artifacts, and `govctl conformance trace` shows which check backs a requirement. `govctl agent install claude` or `govctl agent install codex` installs skills, hooks and reviewers: `/discuss`, `/gov`, `/quick`, `/commit`. For existing projects `/migrate` gradually finds undocumented decisions and records them as ADRs.
Who it is for. For teams and solo developers who delegate code to Claude Code or Codex and want a reviewable record of decisions and conditions for finishing work.
Good fit when
- An idea should not go straight into code: RFC and ADR first, then a work item
- Decisions get lost in chat history and should live in the repository next to the code
- Finishing work should depend on running checks such as clippy or tests
- An old project should be documented gradually: undocumented decisions become ADRs
- An RFC requirement needs to be tied to a test scenario and a check
Not a fit when
- The project is small or throwaway: RFCs and ADRs add more paper than value
- You need a web UI or an MCP server: the author deliberately goes without them and works through the CLI and files
- You use another agent: the ready skills and hooks are built for Claude Code and Codex
- You want conformance cases to run tests themselves: they only declare the link to a check
Example request
/discuss Caching strategy: draft an RFC and an ADR on choosing the cache backend, then create a work itemLimitations
Ready agent integration exists for Claude Code and Codex. Hooks add compact context only in governed projects and advise rather than block direct edits of governance artifacts. Conformance cases declare traceability and do not run tests themselves. Search builds a disposable local index in `.govctl/`, and the TOML files remain the source of truth. Building from source needs Rust and `cargo`, and `cargo binstall` fetches prebuilt binaries. Expensive checks belong only on the work items that need them, with the universal default set kept short.
How to disable. Remove the agent plugin through its plugin manager, the binary with `cargo uninstall govctl`, and drop the `gov/` and `.govctl/` folders from the repository if governance is no longer needed.
FAQ
How do I connect govctl to Claude Code and Codex?
For Claude Code run `govctl agent doctor claude`, then `govctl agent install claude`. For Codex do the same with `codex`. After upgrading govctl run `govctl agent update` and start a new agent session.
How is an RFC different from an ADR in govctl?
An RFC states what must be true: behaviour and constraints. An ADR records why a particular design was chosen. Work items execute against both.
What are verification guards?
They are executable checks that run when a work item completes. The shared set is defined in `gov/config.toml` through `default_guards`, expensive checks attach to specific items, and a guard can be waived with a stated reason.
Does govctl need an MCP server?
No. Artifacts are TOML files in `gov/`, and agents work with them through `list`, `show`, `get` and `edit`. The author deliberately does not hide governance behind a web app or MCP.
How do I adopt govctl in a project already under way?
Run `govctl init`, then `/migrate` in the agent. It finds undocumented decisions and records them as ADRs without restarting the project.
Discussion
No questions or reviews yet.
Related
CLI toolsA Rust SDK for coding agents and the ready-made abstract CLI: REPL, 34 tools, MCP, graph memory, Anthropic and OpenAI, JSON output for CI
A CLI that keeps agent instructions in one place and distributes them to the right config files for Claude Code, Cursor, Codex, Copilot and two dozen other tool
Work modes for AI coding agents: a light solo mode for daily tasks and /man with a plan, checks and review for big changes
agr is a package manager for agent skills: installs them from any Git repo into Claude Code, Cursor and Codex by a versioned manifest, like npm or uv
A coding agent for the terminal and desktop: it does not count "done" without tests or a diff and works with DeepSeek, GLM, Kimi, Claude, Codex and Gemini
A Grid Dynamics plugin that loads a team's shared engineering standards, architecture and workflows into a coding agent session
Comparison with similar tools
Entries that share the main search query, and how they differ.
README in short
The README explains the idea: replace the chain of prompt, code and arguments with RFC or ADR, work item, then code under checks. It then gives a quick start with `cargo install`, agent commands, artifact examples, search, execution loops, guards, conformance cases and migration of existing repositories.