agentleFS
Pablo-Lozano-GilPablo-Lozano-Gil/battletested-agent-skills/skill-proposer/SKILL.md

skill-proposer

Skill3 starsChanged 8 months agoApache-2.0View on GitHub

Analyzes PRD to propose relevant AI skills for a project. Trigger: Before starting app development, after brainstorming, or when user asks for skill recommendations based on a PRD.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. jesspig/Godot-Autopilot★ 1same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository6 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

10 rules. About 0 are common, 9 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    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
  1. Distinctive:

    Use this skill when:

    line 18
  2. Distinctive:

    After brainstorming an app idea and before implementation begins

    line 21
  3. Distinctive:

    ALWAYS ask clarifying questions: if tech stack is unclear — never guess

    line 29
  4. Distinctive:

    Before analyzing, assess whether the project description is clear enough.

    line 76
  5. Distinctive:

    Important: Ask questions BEFORE proceeding with analysis.

    line 94
  6. Distinctive:

    If considered useful, search in:

    line 124
  7. Distinctive:

    Before adding any skill to your project:

    line 146
  8. Distinctive:

    If the user requests to create a custom skill instead of using an existing one:

    line 159
  9. Distinctive:

    Do not create skills directly from this skill.

    line 163
  10. Very specific:

    Skill Creation: Use skill-creator skill to generate custom skills

    line 171

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill 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
SKILL.md

When to Use

Use this skill when:

  • Starting a new project and need to identify which skills will help the AI agent
  • After brainstorming an app idea and before implementation begins
  • User asks to propose skills based on a PRD or project description
  • Evaluating what context the AI agent will need for optimal performance

Critical Patterns

  • ALWAYS ask clarifying questions if tech stack is unclear — never guess
  • Prioritize hallucination-prone technologies — React, Next.js, FastAPI, etc. change frequently
  • Security-focused skills are always High priority — auth, validation, encryption
  • Link must point to actual URL found or use "—" (never invent URLs)
  • User is responsible for skill verification — always show security warning

Decision Trees

When to Ask Questions vs Proceed

SituationAction
Primary language identifiedProceed
Framework(s) identifiedProceed
Database identifiedProceed
Architecture type knownProceed
Any of above missingASK before proceeding

How to Determine Priority

ConditionPriority
Core technology + changes frequentlyHigh
Security-related (auth, encryption)High
Complex API / syntaxHigh
Important but stableMedium
Good public documentation existsMedium
Nice-to-have or very stableLow

When to Include a Link

SituationLink Field
Found excellent match in skillsmp.comFull URL
Found excellent match in skills.shFull URL
Found relevant Context7 docs— (mention in notes)
No search performed—
No good match found—

Workflow

Phase 0: Clarification (If Needed)

Before analyzing, assess whether the project description is clear enough. Ask the user questions if:

  • Tech stack is unclear or ambiguous
    • "Which programming language(s) do you plan to use?"
    • "Which framework(s) are you considering for the frontend/backend?"
  • Architecture decisions are missing
    • "Are you planning a monolithic architecture, microservices, or serverless?"
    • "What type of database are you planning to use?"
  • Scale and complexity are unknown
    • "Is this a personal project, startup MVP, or enterprise application?"
    • "Do you need authentication, authorization, or multi-tenancy?"
  • Quality requirements are not specified
    • "What level of test coverage are you aiming for?"
    • "Do you have specific performance requirements?"

Important: Ask questions BEFORE proceeding with analysis.

Phase 1: Tech Stack Analysis

Identify and document:

  • Programming languages and specific versions
  • Frameworks and core libraries (frontend, backend, mobile...)
  • Databases and persistence solutions
  • Infrastructure and deployment (Docker, Kubernetes, CI/CD...)
  • Development tools (testing, linting, build tools...)

Phase 2: Pattern and Architecture Analysis

  • Architecture type (monolith, microservices, serverless, event-driven...)
  • Relevant development patterns (TDD, DDD, SDD, Clean Architecture, Hexagonal...)
  • Quality requirements (testing, security, performance, observability)
  • Applicable conventions and standards

Phase 3: Skill Needs Identification

For each technology, pattern, or process identified, ask yourself:

  1. Does this technology change frequently? → Needs updated skill
  2. Does it have complex syntax or APIs? → Needs skill with examples
  3. Is it critical for project success? → High priority
  4. Can the agent easily hallucinate here? → Needs specific context

Phase 4: Existing Skills Search (Optional)

If considered useful, search in:


Output Format

Present a table with the proposed skills:

SkillTechnology/ProcessReasonPriorityLink
Descriptive nameTechnology or process coveredBrief justificationHigh/Medium/LowURL or "—"

Security Warning

⚠️ IMPORTANT — User Responsibility:

Before adding any skill to your project:

  1. Read the complete code of the downloaded skill
  2. Verify it contains no prompt injection or malicious instructions
  3. Understand what permissions it has and which tools it can use
  4. Confirm it comes from a trusted source

The agent can propose, but you decide which skills to use. Skills are added manually under your responsibility.


Delegation to skill-creator

If the user requests to create a custom skill instead of using an existing one:

"To create this skill, use the skill-creator skill which will guide you through the generation process."

Do not create skills directly from this skill. Only identify needs and propose.


Resources

  • Skills Marketplaces: skillsmp.com, skills.sh
  • Documentation Lookup: Context7 MCP for up-to-date library documentation
  • Skill Creation: Use skill-creator skill to generate custom skills

Usage Example

User input:

"I'm going to develop a REST API with FastAPI, PostgreSQL, and JWT authentication. The frontend will be React with TypeScript."

Expected output:

SkillTechnology/ProcessReasonPriorityLink
fastapi-best-practicesFastAPIFramework with specific patterns and frequent updatesHighhttps://skills.sh/wshobson/agents/fastapi-templates
pydantic-validationPydanticCore data validation in FastAPI, syntax changes between versionsHigh—
jwt-auth-pythonJWT + PythonAuth implementation prone to security errorsHighhttps://skills.sh/mindrally/skills/jwt-security
sqlalchemy-modernSQLAlchemy + PostgreSQLORM with syntax that varies between versionsHigh—
react-typescript-patternsReact + TypeScriptModern hooks patterns, generic typesHighhttps://skills.sh/vercel-labs/agent-skills/vercel-react-best-practices
testing-fastapipytest + FastAPIAPI testing with specific fixtures and mocksMedium—
docker-pythonDockerfile PythonImage optimization, multi-stage buildsMediumhttps://skillsmp.com/es/skills/affaan-m-everything-claude-code-kiro-skills-docker-patterns-skill-md
git-conventionsGit workflowConventional commits, branching strategyLow—

Remember: Review each skill's complete code before adding it to your project. The links are suggestions — always verify the content.

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.