agent-skills
AGENTS.md2 starsChanged 8 months agoMITView on GitHub
This repository is the canonical source of cross-repo Agent Skills used across every mazrean project. Skills are referenced by apm from mazrean/apm-plackage/* and individual repositories' apm.yml. A skill directory is referenced from a consumer repo's apm.yml like: - Use the creating-agent-skills skill in this repo as the authoring guide. - Each SKILL.md MUST start with frontmatter containing description: (used by Claude Code / apm for auto-discovery). Keep the description concrete — say what the skill does and when to invoke…
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 31,700 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 repository29 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- building-lit-componentsSkill
- building-static-site-rss-feedsSkill
- building-templ-componentsSkill
- checking-wsh-vrtSkill
- committing-codeSkill
- creating-agent-skillsSkill
- creating-test-casesSkill
- deploying-with-isucon-ansibleSkill
- designing-isucon-architectureSkill
- frontend-designSkill
- operating-isucon-servers-via-sshSkill
- planning-wsh-improvementsSkill
- porting-isucon-rust-to-goSkill
- releasing-zig-with-goreleaserSkill
- running-isucon-as-an-agentSkill
- sharing-sockets-with-so-reuseport-in-zigSkill
- tuning-isucon-linux-and-nginxSkill
- using-go-tool-directiveSkill
- using-grafana-mcpSkill
- using-unocss-with-templSkill
- winning-isuconSkill
- winning-web-speed-hackathonSkill
- writing-feature-specSkill
- writing-implementation-tasksSkill
- writing-isucon-arch-subagent-tasksSkill
- writing-project-constitutionSkill
- writing-technical-designSkill
- writing-zig-cli-toolsSkill
The file
We split this AGENTS.md into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
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 28
Use the creating-agent-skills skill in this repo as the authoring guide.
≈ “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
Repository Guidelines — agent-skills
This repository is the canonical source of cross-repo Agent Skills used across
every mazrean project. Skills are referenced by apm
from mazrean/apm-plackage/* and individual repositories' apm.yml.
Layout
skills/
<skill-name>/
SKILL.md # description + frontmatter (auto-discovery metadata)
[reference/*.md] # optional supporting reference files
<single-file>.md # short single-file skills are also allowedA skill directory is referenced from a consumer repo's apm.yml like:
dependencies:
apm:
- mazrean/agent-skills/skills/committing-code
- mazrean/agent-skills/skills/writing-feature-specAuthoring
- Use the
creating-agent-skillsskill in this repo as the authoring guide. - Each
SKILL.mdMUST start with frontmatter containingdescription:(used by Claude Code / apm for auto-discovery). Keep the description concrete — say what the skill does and when to invoke it. - Reference files (loaded only on demand) belong in the skill directory; SKILL.md links to them.
- Single-file skills (e.g.
web-performance-optimization.md) are acceptable when the content fits in one file.
Categories
writing-*— spec-driven development (PRD, design, tasks, constitution). These replacecc-sddandgithub/spec-kitorg-wide.building-*— framework / library tutorials (Lit, templ, RSS feeds, etc.).using-*— toolchain idioms (Go tool directive, UnoCSS-with-templ).creating-*— meta skills (creating new skills, creating test cases).committing-code,releasing-zig-with-goreleaser— workflow skills.web-performance-optimization,frontend-design— design / perf reference.winning-web-speed-hackathon,checking-wsh-vrt,planning-wsh-improvements,sharing-sockets-with-so-reuseport-in-zig,writing-zig-cli-tools— domain references.running-isucon-as-an-agent(contest-day operating manual for agents) andtuning-isucon-linux-and-nginx(transferable infra levers; its prebuilt nginx binary is published as a GitHub Release, not committed) — ISUCON.
Conventions
- Skills must NOT contain repo-specific information (file paths, internal URLs).
Repo-specific guidance belongs in the consumer's
AGENTS.mdor inmazrean/apm-plackage/<stack>/.apm/instructions/. - Bump the
version:field in skill SKILL.md frontmatter (when present) on every behavioural change so consumers can pin. - New skills used by 2+ repos belong here, not in individual repos.
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.