agentleFS
lemieuxlemieux/rfc-skills/CLAUDE.md

rfc-skills

CLAUDE.md9 starsChanged 5 months agoMITView on GitHub

Claude Code skills for RFC and technical proposal writing. Skills follow the Anthropic Agent Skills specification and work with Claude Code (plugin system), Claude.ai (direct upload), and the Claude API. Agents are subagent definitions invoked by skills. They are not user-facing. Quality Guidelines: - Avoid AI-sounding patterns: em dashes, hedging ("It's worth noting"), rhetorical questions, bullet-heavy structure - Write direct, professional prose with active voice - Capture the WHY, not just what - Assume amnesia: write for someone with zero…

The file

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

5 rules. About 0 are common, 2 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Create folder: skills/<skill-name>/ (hyphen-case, lowercase)

    line 57
  2. Very specific:

    Create SKILL.md with required frontmatter:

    line 58
  3. Very specific:

    Add to .claude-plugin/marketplace.json plugins array to register

    line 72
  4. Distinctive:

    Avoid AI-sounding patterns: em dashes, hedging ("It's worth noting"), rhetorical questions, bullet-heavy structure

    line 77
  5. Distinctive:

    Write direct, professional prose with active voice

    line 78

≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.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
CLAUDE.md

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Repository Purpose

Claude Code skills for RFC and technical proposal writing. Skills follow the Anthropic Agent Skills specification and work with Claude Code (plugin system), Claude.ai (direct upload), and the Claude API.

Repository Structure

rfc-skills/
├── CLAUDE.md
├── README.md
├── .claude-plugin/
│   └── marketplace.json        # Plugin marketplace registration
├── skills/
│   ├── brainstorming-rfc/
│   │   └── SKILL.md
│   ├── writing-technical-docs/
│   │   ├── SKILL.md
│   │   └── references/
│   │       ├── rfc-example.md
│   │       ├── rfc-style-guide.md
│   │       ├── rfc-reviewer-prompt.md
│   │       └── rfc-template.md
│   ├── finalizing-technical-docs/
│   │   └── SKILL.md
│   └── incorporating-rfc-feedback/
│       └── SKILL.md
├── agents/                     # Subagent definitions invoked by skills
│   ├── rfc-writer.md
│   └── rfc-reviewer.md
├── spec/                       # Agent Skills specification docs
└── template/
    └── SKILL.md                # Starter template for new skills

Skills Overview

SkillPurpose
brainstorming-rfcPre-RFC exploration and ideation dialogue
writing-technical-docsRFC/technical doc writing and review workflow
finalizing-technical-docsMechanical cleanup before publishing
incorporating-rfc-feedbackProcess and apply reviewer feedback

Agents

Agents are subagent definitions invoked by skills. They are not user-facing.

  • rfc-writer - Writes and updates RFC drafts following the style guide. Handles initial drafting from source material and revisions based on feedback. Self-checks against the style guide before returning; the main agent runs the write→review→revise loop (subagents cannot spawn subagents).
  • rfc-reviewer - Reviews RFC drafts for style and quality issues. Outputs a structured list of issues (does not fix them). Checks for choppy writing, undefined references, missing diagrams, and other quality problems.

Creating a New Skill

  1. Create folder: skills/<skill-name>/ (hyphen-case, lowercase)
  2. Create SKILL.md with required frontmatter:
---
name: skill-name          # Must match folder name
description: |            # When/why to use this skill
  Description here
---

# Skill Instructions
[Markdown body...]
  1. Optional frontmatter: license, allowed-tools, metadata
  2. Add to .claude-plugin/marketplace.json plugins array to register

Skill Design Patterns

Quality Guidelines:

  • Avoid AI-sounding patterns: em dashes, hedging ("It's worth noting"), rhetorical questions, bullet-heavy structure
  • Write direct, professional prose with active voice
  • Capture the WHY, not just what
  • Assume amnesia: write for someone with zero context

Skill Types:

  • Workflow skills (brainstorming-rfc): Guide discovery dialogues, produce structured outputs
  • Style skills (writing-technical-docs): Enforce patterns and anti-patterns via checklists

No Build System

This is a documentation-only repository. No compilation, testing, or linting infrastructure exists. Skills are pure markdown instruction files.

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.