ohmydiff
AGENTS.md0 starsChanged 4 months agoMITView on GitHub
OhMyDiff is a natively discoverable intent-to-diff Agent Skill that also supports direct /omd invocation. skills/omd/SKILL.md is the only behavioral entry point; intent, comparison, and reporting rules live in skills/omd/references/. Codex metadata is in .codex-plugin/ and skills/omd/agents/. .cursor-plugin/ and .claude-plugin/ are unverified packaging previews. Validation scripts live in scripts/, deterministic contracts in tests/, scenarios in evals/, and public architecture in docs/architecture.md. reference/ contains ignored comparison clones. Learn from their implementation patterns, but check licenses and record substantial reuse in NOTICE.md. Before…
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 33,315 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repositoryOne other file it gives its agents.
- omdSkill
The file
We split this AGENTS.md into the rules it gives an agent
3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 15
Use Node.js 22+ with npm and the committed lockfile:
- Distinctive: line 25
Run git diff --check before committing.
- Very specific: line 37
Use atomic Conventional Commits such as refactor: tighten hidden-impact analysis or test: cover conflicting intent sources.
≈ “Common” is an estimate: at least 1 in 5 of the 42,283 AGENTS.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
Repository Guidelines
Project Structure & Module Organization
OhMyDiff is a natively discoverable intent-to-diff Agent Skill that also supports direct /omd invocation. skills/omd/SKILL.md is the only behavioral entry point; intent, comparison, and reporting rules live in skills/omd/references/. Codex metadata is in .codex-plugin/ and skills/omd/agents/. .cursor-plugin/ and .claude-plugin/ are unverified packaging previews. Validation scripts live in scripts/, deterministic contracts in tests/, scenarios in evals/, and public architecture in docs/architecture.md.
reference/ contains ignored comparison clones. Learn from their implementation patterns, but check licenses and record substantial reuse in NOTICE.md.
Reference-First Workflow
Before changing a manifest, skill, evaluation, installer, or contributor rule, inspect the closest implementation in reference/ and current official host documentation. Reuse a proven shape when it fits and document necessary divergence. Never infer a host contract from naming or copy product identity and tuned behavior.
Build, Test, and Development Commands
Use Node.js 22+ with npm and the committed lockfile:
npm ci # install the locked environment
npm run check # run source, link, skill, metadata, and test checks
npm test # run node:test contracts
npm run eval:behavior -- --dry-run # validate the 48-run behavior plan
npm run eval:routing -- --dry-run # validate the routing corpusRun git diff --check before committing.
Coding Style & Naming Conventions
Keep behavior host-neutral and dependency-free. Use two-space indentation for JavaScript, JSON, and YAML. Write ESM .mjs scripts with Node standard-library APIs. Name tests *.test.mjs and fixtures with descriptive kebab-case names. Do not duplicate skill prose in manifests or add runtime code without a demonstrated product need.
Testing Guidelines
Every behavior change needs a failing scenario and expected discrepancy-report contract. Test intent extraction, drift categories, evidence mapping, review hotspots, false positives, and the one-question limit—not exact prose. Codex is the only v0.1 behavioral target. Cursor and Claude remain packaging previews until clean-session acceptance evidence exists.
Commit & Pull Request Guidelines
Use atomic Conventional Commits such as refactor: tighten hidden-impact analysis or test: cover conflicting intent sources. Each commit must remain green and contain one reviewable concern. Pull requests should link the motivating issue, show before-and-after evidence, list host/model versions, disclose AI assistance, and state limitations.
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.

