mcp-operations — Claude Code skill for MCP setup
The mcp-operations skill guides local MCP server checks, host configuration, and safety settings. Use it when tool connectivity fails or MCP behavior changes.
MIT ★ 1,198 updated 2026-10-23
npx skills add hoangsonww/Claude-Code-Agent-Monitor --skill mcp-operations -a claude-codeWhat mcp-operations does
The skill lays out an MCP operations sequence: check /api/health and the server build, review environment flags for mutating or destructive modes, then verify the host configuration path and command. It covers host setup and troubleshooting tool connections.
For code changes, run npm run mcp:typecheck, npm run test:mcp, and npm run mcp:build. Add tools through mcp/src/tools/domains/ and registerAllTools; when a domain changes, update tool-domain-map.md. Read-only operation is the default. Destructive mode is temporary and requires an explicit confirmation token, so enabling it without a specific admin task conflicts with the stated policy.
What it helps with
- check whether the dashboard API responds at
/api/health - trace a failed connection between the host and MCP tools
- review flags that permit mutations or destructive actions
- register an MCP tool in its domain
Example prompts for the agent
- Check the MCP host connection and tell me what to verify first.
- Review this planned change to an MCP tool domain against the project rules.
- Run the required typecheck, MCP tests, and build checks for my changes.
How to install mcp-operations
1Skills CLI (Vercel)
npx skills add hoangsonww/Claude-Code-Agent-Monitor --skill mcp-operations -a claude-code2Manually: put the files in place
~/.claude/skills/mcp-operations/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/mcp-operations/
1Skills CLI (Vercel)
npx skills add hoangsonww/Claude-Code-Agent-Monitor --skill mcp-operations -a codex2Manually: put the files in place
.agents/skills/mcp-operations/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add hoangsonww/Claude-Code-Agent-Monitor --skill mcp-operations -a cursor2Manually: put the files in place
.cursor/skills/mcp-operations/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install hoangsonww/Claude-Code-Agent-Monitor .claude/skills/mcp-operations2Manually: put the files in place
.github/skills/mcp-operations/SKILL.md1Skills CLI (Vercel)
npx skills add hoangsonww/Claude-Code-Agent-Monitor --skill mcp-operations -a gemini-cli2Platform CLI
gemini skills install https://github.com/hoangsonww/Claude-Code-Agent-Monitor --path .claude/skills/mcp-operationsManually: put the files in place
.devin/skills/mcp-operations/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about mcp-operations
How do I install mcp-operations in Claude Code?
Run npx skills add hoangsonww/Claude-Code-Agent-Monitor --skill mcp-operations -a claude-code. The skill file is placed at ~/.claude/skills/mcp-operations/SKILL.md.
What does mcp-operations do?
It provides a workflow for checking the local MCP server, configuring its host, and handling safety flags. For code changes, it calls for typecheck, test, and build commands.
When can destructive MCP mode be enabled?
Only temporarily for an explicit admin task, with a confirmation token. The default policy is read-only.
Source text
SKILL.md · MIT
--- name: mcp-operations description: Operate and maintain the local MCP server for this project. Use when creating MCP host config, troubleshooting tool connectivity, modifying tool domains, or adjusting safety policy flags. --- # MCP Operations Use this skill whenever work touches `mcp/` behavior or MCP host integration. ## Core workflow - Confirm dashboard API is running (`/api/health`). - Confirm MCP server build status. - Validate env flags for mutation/destructive modes. - Verify host configuration path and command. ## Safe operations policy - Default to read-only mode (`MCP_DASHBOARD_ALLOW_MUTATIONS=false`). - Enable mutations only for explicit admin tasks. - Enable destructive mode only transiently and require explicit confirmation token. ## Required verification for code changes - `npm run mcp:typecheck` - `npm run test:mcp` - `npm run mcp:build` ## Catalog invariant - Add tools only through `mcp/src/tools/domains/` and `registerAllTools`. - Protocol transports and REPL must use the same declarations and schemas. - Update `.agents/skills/mcp-maintainer/references/tool-domain-map.md` when a domain changes. ## References - Host config examples: `references/host-config.md` - Operations runbook: `references/runbook.md`
© hoangsonww · MIT · hoangsonww/Claude-Code-Agent-Monitor