Skip to content

parcadei/Continuous-Claude-v3 / .claude/skills/skill-developer

skill-developer

Unverified

Meta-skill for creating and managing Claude Code skills

Skill Developer

Meta-skill for creating new Claude Code skills, including skills that wrap MCP pipelines.

When to Use

  • "Create a skill for X"
  • "Help me make a new skill"
  • "Turn this script into a skill"
  • "How do I create a skill?"

Skill Structure

Skills live in .claude/skills/<skill-name>/:

.claude/skills/my-skill/
├── SKILL.md          # Required: Main skill definition
├── scripts/          # Optional: Supporting scripts
└── templates/        # Optional: Templates, examples

SKILL.md Format

---
name: skill-name
description: Brief description (shown in skill list)
allowed-tools: [Bash, Read, Write]  # Optional: restrict tools
---

# Skill Name

## When to Use
[When Claude should discover this skill]

## Instructions
[Step-by-step instructions for Claude to follow]

## Examples
[Usage examples]

Creating an MCP Pipeline Skill

To create a new MCP chain script and wrap it as a skill:

Step 1: Use the Template

Copy the multi-tool-pipeline template:

cp $CLAUDE_PROJECT_DIR/scripts/multi_tool_pipeline.py $CLAUDE_PROJECT_DIR/scripts/my_pipeline.py

Reference the template pattern:

cat $CLAUDE_PROJECT_DIR/.claude/skills/multi-tool-pipeline/SKILL.md
cat $CLAUDE_PROJECT_DIR/scripts/multi_tool_pipeline.py

Step 2: Customize the Script

Edit your new script to chain the MCP tools you need:

async def main():
    from runtime.mcp_client import call_mcp_tool
    args = parse_args()

    # Chain your MCP tools (serverName__toolName)
    result1 = await call_mcp_tool("server1__tool1", {"param": args.arg1})
    result2 = await call_mcp_tool("server2__tool2", {"input": result1})

    print(result2)

Step 2: Create the Skill

Create .claude/skills/my-pipeline/SKILL.md:

---
name: my-pipeline
description: What the pipeline does
allowed-tools: [Bash, Read]
---

# My Pipeline Skill

## When to Use
- [Trigger conditions]

## Instructions

Run the pipeline:

\`\`\`bash
uv run python -m runtime.harness scripts/my_pipeline.py --arg1 "value"
\`\`\`

### Parameters
- `--arg1`: Description

## MCP Servers Required
- server1: For tool1
- server2: For tool2

Step 3: Add Triggers (Optional)

Add to .claude/skills/skill-rules.json:

{
  "skills": {
    "my-pipeline": {
      "type": "domain",
      "enforcement": "suggest",
      "priority": "medium",
      "description": "What it does",
      "promptTriggers": {
        "keywords": ["keyword1", "keyword2"],
        "intentPatterns": ["(pattern).*?(match)"]
      }
    }
  }
}

Reference Files

For full details, read:

cat $CLAUDE_PROJECT_DIR/.claude/rules/skill-development.md
cat $CLAUDE_PROJECT_DIR/.claude/rules/mcp-scripts.md

Quick Checklist

  • SKILL.md has frontmatter (name, description)
  • "When to Use" section is clear
  • Instructions are copy-paste ready
  • MCP servers documented if needed
  • Triggers added to skill-rules.json (optional)

Examples in This Repo

Look at existing skills for patterns:

ls $CLAUDE_PROJECT_DIR/.claude/skills/
cat $CLAUDE_PROJECT_DIR/.claude/skills/commit/SKILL.md
cat $CLAUDE_PROJECT_DIR/.claude/skills/firecrawl-scrape/SKILL.md
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 463kUpdated Claude CodeCodexVerifiedInstall
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 463kUpdated Claude CodeCodexVerifiedInstall
discernment-nudgeanthropics/skillsAfter you give a substantive answer or draft that the user may act on — advice or recommendations, drafted artifacts such as goals, plans, pitches, proposals, or emails, estimates or projections, analysis or interpretation of data, factual claims they may rely on, or a multi-step argument — invoke this skill BEFORE finalizing your reply and then, if it applies, append 2-3 short follow-up questions, each tied to something specific in what you just produced, that help the user check key facts, probe the reasoning or assumptions, and notice missing context. Do this at most once per conversation. Skip it when the user asked a trivial how-to or simple lookup, wants a purely educational explanation, asked you only to format, convert, or assemble a file from content they provided, is writing code they will run, is doing creative writing or casual chat, or already asked you to double-check, cite, or review — the skill file explains these boundaries and the exact output format.Stars: 195kUpdated Claude CodeCodexVerifiedInstall
internal-commsanthropics/skillsA set of resources to help me write all kinds of internal communications, using the formats that my company likes to use. Claude should use this skill whenever asked to write some sort of internal communications (status reports, leadership updates, 3P updates, company newsletters, FAQs, incident reports, project updates, etc.).Stars: 195kUpdated Claude CodeCodexVerifiedInstall