lean-spec
CLAUDE.md296 starsChanged 6 months agoMITView on GitHub
This file defines shared conventions enforced across all codervisor repositories. It is synced from the codervisor meta-repo into each child project. codervisor is a suite of AI-agent infrastructure projects: Types: feat, fix, refactor, test, docs, chore, ci, perf. Examples: feat/session-pool, fix/auth-timeout, chore/update-deps.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- codervisor/leanspecword for word, first seen May 20, 2026
Other versions1 project uses a version of this.
Word for word · 1
- codervisor/leanspec★ 294100% same
Edited · 0
None found.
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository8 other files it gives its agents.
- AGENTS.mdAGENTS.md
- docs-site/AGENTS.mdAGENTS.md
- leanspec-developmentSkill
- leanspec-dev-processSkill
- leanspec-pre-pushSkill
- leanspec-pr-lifecycleSkill
- watch-ciSkill
- leanspecSkill
The file
Show its rules and text
We split this CLAUDE.md into the rules it gives an agent
13 rules. About 0 are common, 9 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈9
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- could be automatic: a setting could enforce it
- Distinctive: line 23
Write clear, self-documenting code.
- Distinctive: line 24
Prefer small, focused commits with descriptive messages (imperative mood, <72 chars).
- Distinctive: line 25
Every PR must pass CI before merge.
- Distinctive: line 26
Keep dependencies minimal.
- Very specific: line 33
Use thiserror for library errors, anyhow for binary/application errors.
- Very specific: line 34
Prefer #[must_use] on functions returning values that should not be silently dropped.
- Very specific: line 42
Use named exports. Avoid default exports except for framework conventions.
- Very specific: line 43
Prefer async/await over raw Promises.
- Distinctive: line 48
Follow Mathlib conventions for naming and style.
- Distinctive: line 49
Keep proofs tactic-mode where possible for readability.
- Distinctive: line 73
Keep PRs small and focused.
- Distinctive: line 78
Never commit secrets, tokens, or credentials.
- Distinctive: line 80
Follow OWASP top-10 awareness for any web-facing code.
≈ “Common” is an estimate: at least 1 in 5 of the 28,029 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
CLAUDE.md — codervisor project standards
This file defines shared conventions enforced across all codervisor repositories. It is synced from the codervisor meta-repo into each child project.
Repository overview
codervisor is a suite of AI-agent infrastructure projects:
| Repo | Stack | Purpose |
| stiglab | Rust + TypeScript | Distributed AI agent session orchestration |
| synodic | TypeScript | AI harness governance framework |
| telegramable | TypeScript | Telegram-first AI agent proxy |
| ising | Rust | Code graph analysis engine |
| skills | TOML / Markdown | Shared Claude Code skills |
| lean-spec | Lean 4 | Formal specification framework |
Coding conventions
General
- Write clear, self-documenting code. Add comments only where intent is non-obvious.
- Prefer small, focused commits with descriptive messages (imperative mood, <72 chars).
- Every PR must pass CI before merge. No force-pushing to
main. - Keep dependencies minimal. Justify new crates / packages in the PR description.
Rust repos (stiglab, ising)
- Edition: 2021 or later.
- Format with
rustfmt(default config unlessrustfmt.tomlis present). - Lint with
clippy— treat warnings as errors in CI (-D warnings). - Use
thiserrorfor library errors,anyhowfor binary/application errors. - Prefer
#[must_use]on functions returning values that should not be silently dropped. - Tests live next to the code in
#[cfg(test)]modules; integration tests intests/.
TypeScript repos (synodic, telegramable)
- Target: ES2022+ / Node 20+.
- Strict mode:
"strict": trueintsconfig.json. - Lint with Biome (preferred) or ESLint. Format with Biome or Prettier.
- Use named exports. Avoid
defaultexports except for framework conventions. - Prefer
async/awaitover raw Promises. Avoid.then()chains. - Tests use Vitest (preferred) or Jest. Co-locate test files as
*.test.ts.
Lean repos (lean-spec)
- Follow Mathlib conventions for naming and style.
- Keep proofs tactic-mode where possible for readability.
Commit messages
<type>: <short summary in imperative mood>
Optional body explaining *why*, not *what*.Types: feat, fix, refactor, test, docs, chore, ci, perf.
Branch naming
<type>/<short-description>Examples: feat/session-pool, fix/auth-timeout, chore/update-deps.
PR standards
- Title follows commit message format.
- Description includes a Summary (what and why) and Test plan (how it was verified).
- Keep PRs small and focused. Split large changes into stacked PRs.
- Request review from at least one team member.
Security
- Never commit secrets, tokens, or credentials. Use environment variables.
- Validate all external input at system boundaries.
- Follow OWASP top-10 awareness for any web-facing code.
- Pin CI action versions to full SHA, not tags.
Dependencies
- Rust: pin exact versions in
Cargo.tomlfor binaries; use semver ranges for libraries. - TypeScript: use a lockfile (
package-lock.jsonorpnpm-lock.yaml). Commit it. - Review dependency updates via Dependabot or Renovate PRs — don't auto-merge.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

