Stats
Actions
Tags
From rpw-building
Use when MCP servers fail due to missing .env, authentication errors, or env var issues. Provides tips for MCP environment setup when .env may be missing or incomplete.
How this skill is triggered — by the user, by Claude, or both
Slash command
/rpw-building:mcp-env-setupThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Tips for setting up MCP server environments when `.env` may be missing or incomplete.
Tips for setting up MCP server environments when .env may be missing or incomplete.
.env file is missing or not foundLocate .env.example — every MCP server in this repo ships one. Copy it:
cp .env.example .env
Fill required variables — open .env and replace placeholders with real values. Never commit .env.
Verify .env location — run_mcp.py typically loads from the server directory. Ensure .env is in the same directory as run_mcp.py or that the launcher's working directory is correct.
Check Cursor MCP config — Cursor reads MCP config from settings. Ensure:
${CLAUDE_PLUGIN_ROOT} or equivalent so they resolve on the user's machine.env at runtimeTest manually — from the MCP server directory:
uv run python run_mcp.py
If it fails, the error usually indicates which env var is missing.
| Symptom | Likely cause | Fix |
|---|---|---|
| "API key not found" | .env missing or wrong path | Copy .env.example to .env, fill values |
| "Invalid credentials" | Wrong or expired token | Re-auth (e.g. sf org login web for Salesforce, OAuth for Google) |
| "Module not found" | Dependencies not installed | uv sync in the MCP server directory |
| MCP shows "disconnected" | Config path wrong | Use ${CLAUDE_PLUGIN_ROOT} placeholders |
Never hardcode machine-specific paths (e.g. /Users/foo/...) in MCP config. Use ${CLAUDE_PLUGIN_ROOT} or env-driven placeholders so configs work across machines.
npx claudepluginhub randypitcherii/rpw-agent-marketplace --plugin rpw-buildingProvides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.