start_here — Agent Discovery Doctor - llms.txt, agent card, mcp.json MCP Prompt
start_here
is one of 1 prompt on the
Agent Discovery Doctor - llms.txt, agent card, mcp.json
MCP server. Connect the server and it appears in your client, ready to invoke.
by client
How to invoke start_here from your client
- Claude Code Agent Discovery Doctor - llms.txt, agent card, mcp.json Prompt/start_here run in your project directory
- Claude Desktop Agent Discovery Doctor - llms.txt, agent card, mcp.json Prompt/start_here ~/Library/Application Support/Claude/claude_desktop_config.json
- Cursor Agent Discovery Doctor - llms.txt, agent card, mcp.json Prompt/start_here ~/.cursor/mcp.json
- VS Code Agent Discovery Doctor - llms.txt, agent card, mcp.json Prompt/start_here .vscode/mcp.json
- Zed Agent Discovery Doctor - llms.txt, agent card, mcp.json Prompt/start_here ~/.config/zed/settings.json
- Windsurf Agent Discovery Doctor - llms.txt, agent card, mcp.json Prompt/start_here ~/.codeium/windsurf/mcp_config.json
- Cline Agent Discovery Doctor - llms.txt, agent card, mcp.json Prompt/start_here ~/Library/Application Support/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json
- Gemini CLI Agent Discovery Doctor - llms.txt, agent card, mcp.json Prompt/start_here ~/.gemini/settings.json
- Grok Agent Discovery Doctor - llms.txt, agent card, mcp.json Prompt/start_here .mcp.json (in your project root)
- ChatGPT Agent Discovery Doctor - llms.txt, agent card, mcp.json Prompt/start_here Settings → Connectors → Advanced → Developer mode
- Claude.ai Agent Discovery Doctor - llms.txt, agent card, mcp.json Prompt/start_here Settings → Connectors → Add custom connector
- LangChain Agent Discovery Doctor - llms.txt, agent card, mcp.json Prompt/start_here pip install langchain-mcp-adapters
Agent Discovery Doctor - llms.txt, agent card, mcp.json also offers 4 resources — data you attach as context, where a prompt is something you run.
Read from the server itself, by connecting to it and calling prompts/list on 24 September 2026.
A listing does not declare its prompts, so asking is the only way to know them, and this
is what the server actually offers rather than what its listing claims. Names only are
stored; connect the server for each prompt's arguments.