udaysharmadevudaysharmadev/Skills/AGENTS.md

Skills

AGENTS.md12 starsChanged 56 days agoMITView on GitHub

This repository is an Agent Skills bundle. Everything in PRD.md applies to you while you work here. Phase 7 of the build sequence means it: use this suite's own methodology (evidence, verification, no slop) on this repository. Anything annoying in real usage gets fixed before launch.

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 repository28 other files it gives its agents.
- backendSkill
- blueprintSkill
- clearedSkill
- conciergeSkill
- distillSkill
- dittoSkill
- findableSkill
- frictionSkill
- frontpageSkill
- frugalSkill
- handsfreeSkill
- hardenSkill
- headroomSkill
- hotpathSkill
- hotseatSkill
- janitorSkill
- masterplanSkill
- pilotSkill
- polishSkill
- proofSkill
- recallSkill
- refereeSkill
- roadtestSkill
- runwaySkill
- scoutSkill
- sleuthSkill
- spelunkSkill
- unslopSkill

The file

We split this AGENTS.md into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 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.
0
Very specific
Tied to their project's names, files or tools. Probably skip.

Common
Distinctive
Very specific

Distinctive:
No slop: No fake benchmark numbers, no invented compatibility claims, no decorative badges, no filler sections in skill files.
line 8

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.

Files

AGENTS.md

AGENTS.md — guidance for agents working in this repository

This repository is an Agent Skills bundle. Everything in PRD.md applies to you while you work here.

Non-negotiables

Run the validators before claiming done. scripts/validate-skills, scripts/check-names, scripts/check-links. A clean-looking tree is not evidence.
No slop. No fake benchmark numbers, no invented compatibility claims, no decorative badges, no filler sections in skill files. If a number is not measured, it does not ship.
Skills are self-contained. A skill folder under skills/ must work when copied alone into another repository. Reference sibling skills by slug name only (e.g. "hand off to masterplan") — never by relative file path into ../. Canonical contracts live in shared/ for humans and tooling, not for runtime cross-loads.
Respect the skill-file contract (PRD §8): every SKILL.md states purpose, triggers, when NOT to use, prerequisites, workflow, tool selection/fallback, quality gates, stop conditions, output contract, and pointers to references. Body ≤ 500 lines; detail goes in references/.
Naming: single lowercase word per skill (PRD §16). scripts/check-names enforces it.
Progressive disclosure: frontmatter description carries the trigger signal; the body carries the workflow; references/ carries the depth. Do not inline checklists that belong in a reference file.
Paid agents are default-deny. The eval harness refuses codex (and any future billed adapter) unless the human sets BOTH --allow-paid and ALLOW_PAID_AGENT=1 on that run — scripts/eval_guard.py exits 4 otherwise. Never set those keys yourself and never invoke a paid agent CLI directly (e.g. codex exec in a shell) unless the user explicitly authorized that spend in the current session.

Before opening a PR

scripts/validate-skills passes
scripts/check-names passes
scripts/check-links passes
New/changed skills have trigger cases in evals/trigger/cases.md
scripts/build-docs regenerated if frontmatter changed

Dogfooding

Phase 7 of the build sequence means it: use this suite's own methodology (evidence, verification, no slop) on this repository. Anything annoying in real usage gets fixed before launch.

Download now (ZIP)

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

What did you use it for? e.g. Reviewing pull requests in a Rails app

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.

Check your setup, free