engineering-review-board
AGENTS.md11 starsChanged 9 days agoMITView on GitHub
This repository is the canonical local source for reusable agent skills. Before editing skills or repository documentation, read README.md. - Edit only first-party skills as repository source. - Treat lockfile-owned or .gitignore-ignored directories under skills/ as third-party runtime installs, not first-party project assets. - Do not force-add ignored third-party skill directories or copy raw third-party artifacts into first-party skills without explicit review. - Keep first-party skill guidance project-neutral and reusable across repositories. - Confirm SKILL.md front matter includes matching name…
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 31,843 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 repository70 other files it gives its agents.
- adversarial-reviewSkill
- api-designSkill
- architecture-reviewSkill
- behavior-driven-developmentSkill
- brainstormingSkill
- ci-release-engineeringSkill
- clean-architectureSkill
- code-reviewSkill
- container-engineeringSkill
- create-agent-skillSkill
- csharp-dotnet-engineeringSkill
- css-scss-stylingSkill
- data-platform-engineeringSkill
- dependency-supply-chain-reviewSkill
- digital-asset-managementSkill
- documentation-engineeringSkill
- domain-driven-designSkill
- domain-modelingSkill
- donsetch-web-researchSkill
- gherkinSkill
- git-commitSkill
- git-workflowsSkill
- hexagonal-architectureSkill
- impeccableSkill
- internationalization-localizationSkill
- javascript-typescript-engineeringSkill
- justfilesSkill
- mcp-server-engineeringSkill
- mysql-mariadb-sql-engineeringSkill
- observability-engineeringSkill
- onion-architectureSkill
- parallelism-engineeringSkill
- performance-reviewSkill
- playwright-e2eSkill
- postgresql-sql-engineeringSkill
- prompt-engineering-reviewSkill
- python-antipatternsSkill
- python-design-patternsSkill
- python-engineeringSkill
- random-data-identifiersSkill
- release-readinessSkill
- review-verification-protocolSkill
- root-cause-analysisSkill
- ruby-engineeringSkill
- rust-antipatternsSkill
- rust-async-webSkill
- rust-code-reviewSkill
- rust-design-patternsSkill
- rust-desktop-guiSkill
- rust-engineeringSkill
- rust-persistence-sqlSkill
- rust-testing-qualitySkill
- script-engineeringSkill
- security-review-evidenceSkill
- security-reviewSkill
- semantic-versioningSkill
- sql-engineeringSkill
- sqlite-sql-engineeringSkill
- suggest-lucide-iconsSkill
- svelte-sveltekit-engineeringSkill
- The first 60 shown.
The file
We split this AGENTS.md into the rules it gives an agent
4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 9
Treat lockfile-owned or .gitignore-ignored directories under skills/ as third-party runtime installs, not first-party project assets.
- Distinctive: line 11
Do not force-add ignored third-party skill directories or copy raw third-party artifacts into first-party skills without explicit review.
- Distinctive: line 16
Keep first-party skill guidance project-neutral and reusable across repositories.
- Distinctive: line 27
Treat external-repository runners as untrusted.
≈ “Common” is an estimate: at least 1 in 5 of the 40,246 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
Agent Instructions
This repository is the canonical local source for reusable agent skills. Before
editing skills or repository documentation, read README.md.
First-Party Skill Rules
- Edit only first-party skills as repository source.
- Treat lockfile-owned or
.gitignore-ignored directories underskills/as third-party runtime installs, not first-party project assets. - Do not force-add ignored third-party skill directories or copy raw third-party artifacts into first-party skills without explicit review.
Skill and Docs Maintenance
- Keep first-party skill guidance project-neutral and reusable across repositories.
- Confirm
SKILL.mdfront matter includes matchingnameand a non-emptydescription, and check local Markdown links after editing a skill.
Runtime and Execution Boundaries
- Host configuration and runtime authority are machine-local; repository skills describe procedures and never grant authority.
- Static permissions may confirm direct command forms but do not sandbox runners, transitive effects, filesystem access, network access, or the host process.
- Treat external-repository runners as untrusted. Use operating-system, container, or VM isolation for genuinely untrusted execution.
- Synthetic stdin, prompts, and Task packets do not protect repository files, ambient credentials, inherited environment, egress, or filesystem access.
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.