fullstack-agent-skills / for-codex
AGENTS.md0 starsChanged 6 months agoMITView on GitHub
This project uses the fullstack-agent-skills pack. Skills live in ./skills/ (or ./.codex-skills/ if symlinked). Each skill is a markdown file describing a focused workflow. When a user request matches one of the triggers below, load the matching skill file and follow its workflow before responding. Do not freelance the work. If no skill matches, proceed normally — but say which skill you considered and why it didn't fit, in one line.
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 31,774 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 repository16 other files it gives its agents.
- api-architectSkill
- browser-qaSkill
- code-auditorSkill
- doc-craftSkill
- git-flow-proSkill
- mcp-forgeSkill
- perf-hunterSkill
- refactor-masterSkill
- resilience-engineerSkill
- schema-architectSkill
- security-sentinelSkill
- ship-itSkill
- test-architectSkill
- threat-modelerSkill
- type-tightenerSkill
- ui-polishSkill
The file
We split this AGENTS.md into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are 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.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 7
When a user request matches one of the triggers below, load the matching skill file and follow its workflow before responding.
- Distinctive: line 32
One skill per task: If the request spans two skills, pick the dominant one and call out the other in your handoff.
- Distinctive: line 40
If no skill matches, proceed normally — but say which skill you considered and why it didn't fit, in one line.
≈ “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
Codex agent instructions — fullstack-agent-skills routing
This project uses the fullstack-agent-skills pack. Skills live in ./skills/ (or ./.codex-skills/ if symlinked). Each skill is a markdown file describing a focused workflow.
How to use a skill
When a user request matches one of the triggers below, load the matching skill file and follow its workflow before responding. Do not freelance the work.
Routing table
| User says (or implies) | Load skill |
| "threat model this", "STRIDE analysis", "what's the attack surface", "what could go wrong here" | skills/threat-modeler/SKILL.md |
| "build an MCP for", "wrap this API as MCP", pastes OpenAPI spec | skills/mcp-forge/SKILL.md |
| "test this flow", "verify the signup", "qa my app", "browser test" | skills/browser-qa/SKILL.md |
| "make this look good", "polish the UI", "improve the design", "looks like AI slop" | skills/ui-polish/SKILL.md |
| "review this code", "audit", "find bugs", "code review" | skills/code-auditor/SKILL.md |
| "refactor this", "clean up", "split this module", "this file is too big" | skills/refactor-master/SKILL.md |
| "write tests for", "cover with tests", "add unit tests" | skills/test-architect/SKILL.md |
| "commit this", "make a PR", "write the commit message", "clean up my branch" | skills/git-flow-pro/SKILL.md |
| "design an API for", "plan the endpoints", "OpenAPI spec for" | skills/api-architect/SKILL.md |
| "design the schema", "model this data", "create a migration", "what indexes do I need" | skills/schema-architect/SKILL.md |
| "deploy this", "ship it", "set up Vercel/Railway/Fly", "dockerize this", "write the deploy action" | skills/ship-it/SKILL.md |
| "write a README", "document this", "add docstrings", "write a CHANGELOG" | skills/doc-craft/SKILL.md |
| "make this faster", "why is this slow", "optimize this", "the query times out" | skills/perf-hunter/SKILL.md |
| "security audit", "check for vulnerabilities", "is this safe", "OWASP" | skills/security-sentinel/SKILL.md |
| "remove any", "enable strict mode", "fix the typescript", "type this" | skills/type-tightener/SKILL.md |
| "improve error handling", "add logging", "make this resilient", "this should retry" | skills/resilience-engineer/SKILL.md |
Operating rules
- One skill per task. If the request spans two skills, pick the dominant one and call out the other in your handoff.
- Follow the workflow as written. Each skill has a numbered workflow. Don't skip steps.
- Output contract matters. Each skill commits to what success looks like — meet the contract before declaring done.
- Anti-patterns are non-negotiable. Each skill has a "Don't" list — those are bugs waiting to happen. Skip them.
- Chain when appropriate. Skills cross-reference each other in their
See alsosections. After completing one, suggest the next if it fits.
Fallback
If no skill matches, proceed normally — but say which skill you considered and why it didn't fit, in one line.
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.