agentleFS
HashLipsHashLips/agent-skills/skills/agent-creator/SKILL.md

agent-creator

Skill29 starsChanged 7 months agoMITView on GitHub

Authors lean, role-driven agent definitions in Markdown (responsibilities, decision framework, constraints, outputs, completion and handoff, collaboration) for use in any project or product. Use when creating `*-agent.md` files, standardizing a portable agent library, or refactoring ad-hoc prompts so they stay scannable and under a reasonable context budget.

26 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use security, skill-creator and 2 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • 0xSeroSkill

    security

    Also in 4 projects that use this

  • SentrySkill

    skill-creator

    Create new agent skills following the Agent Skills specification. Use when asked to "create a skill", "add a new skill", "write a skill", "make a skill", "build a skill", or scaffold a new skill with SKILL.md. Guides through requirements, writing, registration, and verification.

    Also in 10 projects that use this

  • SentrySkill

    security-review

    Finds exploitable application security vulnerabilities in code changes. Use for Warden security scans, appsec review, OWASP-style checks, authentication or authorization bugs, injection, XSS, SSRF, path traversal, secrets, unsafe crypto, webhook verification, open redirects, or sensitive data exposure.

    Also in 3 projects that use this

  • GoogleSkill

    code-review

    Adversarial systems, concurrency, memory-model, and architecture code review protocol. Use this skill whenever asked to review code, PRs, diffs, or branches in Filament.

    Also in 5 projects that use this

The file

We split this skill into the rules it gives an agent

8 rules. 0 are common, 6 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 24 edited versions we read keep these.

  • 6

    Distinctive

    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
  1. Distinctive:

    Keep agents concise while preserving operational clarity.

    line 11
  2. Distinctive:

    Add role summary only if Identity is not enough.

    line 18
  3. Distinctive:

    Add responsibilities, decision framework, constraints, outputs, and collaboration.

    line 19
  4. Very specific:

    Add required Failure modes and recovery micro-section.

    line 20
  5. Very specific:

    Add required Completion and handoff section.

    line 21
  6. Distinctive:

    Add optional sections only when they add signal.

    line 22
  7. Distinctive:

    Run compression pass (dedupe, shorten, reference reusable detail).

    line 23
  8. Distinctive:

    Run hard QA gate before delivery.

    line 24

“Common” is measured: kept by at least 6 in 10 of the 24 edited versions of this file we could read. “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
SKILL.md

Agent Creator

Summary

  • Build role-driven agent files that are portable and scannable.
  • Keep agents concise while preserving operational clarity.
  • Require explicit completion and handoff behavior in every agent.

Core Workflow

  1. Name the file in kebab-case using <role>-agent.md.
  2. Fill Identity as one paragraph from the template.
  3. Add role summary only if Identity is not enough.
  4. Add responsibilities, decision framework, constraints, outputs, and collaboration.
  5. Add required Failure modes and recovery micro-section.
  6. Add required Completion and handoff section.
  7. Add optional sections only when they add signal.
  8. Run compression pass (dedupe, shorten, reference reusable detail).
  9. Run hard QA gate before delivery.
  10. Apply final formatting pass using the available MD Design System skill.

Non-Negotiable Rules

  • File name — Must be <descriptor>-agent.md.
  • Required sections — Responsibilities, decision framework, constraints, failure modes, outputs, collaboration, completion and handoff.
  • Completion contract — Must include DoD, stop condition, next role + package, and start rule.
  • Identity format — Single paragraph only.
  • Readable Markdown — Minimal bold; list-first; no emphasis noise.
  • Brevity target — Roughly 250-500 words per agent.
  • Portability — No mandatory repo-local paths.
  • Skill boundary — Agent defines role behavior; separate skills define reusable procedures.
  • Final formatter — Before final output, run generated files through the local MD design system formatter.

Output Contract

  • Agent file is usable as-is in system instructions.
  • Reader can quickly identify boundaries, outputs, and handoff rules.
  • Next role can start from the handoff package without ambiguity.

Reference Index

  • Completion and handoff requirements: references/completion-and-handoff.md
  • Agent template and section order: references/agent-file-template.md
  • Naming and layout rules: references/naming-and-layout.md
  • Authoring process and review flow: references/authoring-workflow.md
  • Hard pass/fail QA gate: references/qa-rubric.md
  • Failure-mode section patterns: references/failure-modes.md
  • Role-family compact exemplars: references/role-family-exemplars.md
  • Role vs persona distinction: references/role-vs-persona.md
  • Collaboration patterns: references/collaboration-patterns.md
  • Final Markdown formatting contract: available MD Design System skill

When To Use This Skill

  • Creating or rewriting role-based agent files.
  • Standardizing agent libraries across repos.
  • Converting verbose persona prompts into operational role specs.

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.