Mcp server design
Skill jukrap/ai-agent-playbook/skills/ai-harness/mcp-server-design
Use when designing or changing MCP tools, resources, prompts, permission tiers, write gates, cache/index surfaces, or agent harness integrations.From its SKILL.md
npx -y skills add jukrap/ai-agent-playbook --skill mcp-server-designAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
0.9 KB, 152 tokens by cl100k_base, as published. Nobody here has run it
MCP Server Design
Use this as the primary AI harness skill for MCP and integration surfaces.
Workflow
- Classify the capability as resource, prompt, read tool, scaffold tool, managed write, or project write.
- Keep read-only tools default; require explicit opt-in and
applysemantics for writes. - Return structured content, concise text summaries, target path evidence, and clear conflicts.
- Add tests that prove read-only tools do not write and write tools stay disabled by default.
Reference
Read references/runtime-boundaries-and-permissions.md when MCP work includes a daemon, transport, session store, queue mode, tunnel/remote surface, scoped token, approval flow, or server-backed search/read bridge.
What ships with it: 1 file
3.8 KB alongside SKILL.md