Mcp Operations
This handles the operational side of running an MCP server for your project: setting up host configs, debugging connection issues, and managing safety flags. You'll reach for it when you're wiring up MCP integration or when tools aren't connecting properly. The workflow is sensible, defaulting to read-only mode and requiring explicit confirmation before enabling destructive operations. It's essentially the ops runbook packaged as a skill, with built-in guardrails around mutations and a clear verification path through typecheck and build commands. Useful if you're maintaining an MCP server and want consistent operational practices baked in.
- GitHub stars
- 455
- last updated
- Jun 11, 2026
Install
$ npx -y skills add hoangsonww/claude-code-agent-monitor --skill mcp-operations --agent claude-codeInstalls into .claude/skills of the current project.
Files
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:typechecknpm run test:mcpnpm run mcp:build
Catalog invariant
- Add tools only through
mcp/src/tools/domains/andregisterAllTools. - Protocol transports and REPL must use the same declarations and schemas.
- Update
.agents/skills/mcp-maintainer/references/tool-domain-map.mdwhen a domain changes.
References
- Host config examples:
references/host-config.md - Operations runbook:
references/runbook.md
Sponsors
Details
- Repository
- github.com/hoangsonww/claude-code-agent-monitor
- Categories
- Backend & APIs, DevOps & CI/CD, AI & Agent Building, Debugging
- License
- MIT
- First seen
- Jun 11, 2026
