Skip to content

agent-install-prompt.en.md

Latest commit

 

History

History
86 lines (64 loc) · 2.87 KB

agent-install-prompt.en.md

File metadata and controls

86 lines (64 loc) · 2.87 KB

Have an Agent Install fcop-mcp For You (canonical prompt / en)

Send the block below to a fresh agent that can run shell commands (new Cursor chat, or any AI with terminal access). It will install fcop-mcp into Cursor step by step, no manual work from you.

This prompt is the source-of-truth in the fcop repo at src/fcop/rules/_data/agent-install-prompt.en.md, also exposed as the MCP resource fcop://prompt/install — the same text powers the GitHub README, the PyPI project page, and the human-readable instructions for agents.

Copy the block below and send it to your agent

Install fcop-mcp into Cursor for me — you run the commands end to end.

1. Detect my OS first: in the terminal, run
   `uname -s 2>$null; echo $env:OS` to see whether this is Windows
   or macOS / Linux.

2. Install uv (if not already present). One-liner:
   - Windows PowerShell:
     powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"
   - macOS / Linux:
     curl -LsSf https://astral.sh/uv/install.sh | sh
   Then run `uvx --version` to confirm.

3. Add an fcop entry to the global mcp.json. **Preserve** existing
   mcpServers — do not overwrite them.
   - Windows path: %USERPROFILE%\.cursor\mcp.json
   - macOS / Linux path: ~/.cursor/mcp.json
   - Add this snippet inside the mcpServers object:
     "fcop": {
       "command": "uvx",
       "args": ["fcop-mcp"]
     }

4. Print the final mcp.json contents back to me.

5. Remind me to restart Cursor; on first launch fcop-mcp will pull
   dependencies, **wait 30 seconds to 1 minute**, do not close or
   reconnect early.

Report back after each step before moving on. **Do not** auto-init
a project after install — initialization is my (ADMIN's) choice; I
will pick solo / dev-team / custom myself.

After install: run fcop_audit() for a health check

Once fcop-mcp is installed and the project is initialized, have the agent run a health check immediately:

Run fcop_audit(scope="new") and give me a project health check report.

The health check report (INSPECTION-*.md) will tell you:

  • Whether all protocol files are in place
  • Whether any role document gaps exist
  • Any P0 / P1 / P2 items needing remediation

If you're taking over an existing project, use scope="takeover" instead of scope="new".

After install: initialization is ADMIN's pick

Once fcop-mcp is installed, the agent must not default to init_project(team="dev-team") —

  • Is this a 4-person team?
  • Or solo (single AI)?
  • Or a custom roster?

That's ADMIN's call (yours), not the agent's default. After Cursor restarts, in a new chat the agent will first call fcop_report(), whose output includes a 3-way choice. You pick, then the agent can invoke the matching init_* tool.

For the full reasoning, read fcop/LETTER-TO-ADMIN.md in your project — that's fcop's complete manual for ADMIN.