Skill mcp registry management
MCP registry skill — register servers, discover tools, allow-lists, health monitoring
npx -y skills add zavora-ai/skill-mcp-registry-managementAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Manage the MCP server registry — register servers, discover tools, configure allow-lists, monitor health, restart unhealthy servers, and export inventory. Use when onboarding MCP servers, checking available tools, managing permissions, monitoring health, or troubleshooting connectivity.
The file declares its own license as Apache-2.0. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
1.8 KB, 241 tokens by cl100k_base, as published. Nobody here has run it
MCP Registry Management
You manage the MCP server control plane. Register servers, discover tools, enforce allow-lists, and monitor health. No server is accessible by default — explicit allow-listing required.
Decision Tree
├── "what servers", "available", "list"? → list_servers / discover_tools
├── "register", "onboard", "add server"? → register_server
├── "health", "status", "is it up"? → health_check / inspect_server
├── "allow", "permission", "enable tool"? → allow_list_tools
├── "restart", "fix", "unhealthy"? → restart_server
├── "disable", "remove"? → disable_server
├── "export", "inventory"? → export_inventory
MUST DO
- Require manifest validation before registering
- Enforce allow-lists (no default access)
- Monitor health continuously
- Classify all tools by risk level before enabling
MUST NOT DO
- Don't register servers without valid manifests
- Don't enable tools without risk classification
- Don't disable servers without checking dependents
What ships with it: 8 files
10.3 KB alongside SKILL.md, 1 of them executable
assets/
- registry-health.md720 B
references/
- cross-mcp-workflows.md691 B
- examples.md2.4 KB
- tool-sequences.md2.6 KB
scripts/
- health_report.pyruns671 B
- .gitignore29 B
- LICENSE738 B
- README.md2.5 KB
Gives 0 of the 12 instructions most mcp tooling skills give in 241 tokens
Counted across 638 of the 750 authors here whose files we hold, read 2026-08-07
- Create ten complex or independent read-only evaluation questionsin 69 of 638, across 15 files
- Test servers using MCP Inspectorin 61 of 638, across 19 files
- Provide actionable error messages with specific next stepsin 54 of 638, across 12 files
- Prioritize comprehensive API coverage over specific workflows or workflow toolsin 54 of 638, across 12 files
- Use TypeScript and Streamable HTTP for remote servers or clientsin 54 of 638, across 8 files
- Define structured output schemas where possiblein 50 of 638, across 8 files
- Use Zod or Pydantic for input schemasin 47 of 638, across 5 files
- Fetch MCP specification pages with markdown suffixin 46 of 638, across 4 files
- Load framework documentation using WebFetchin 45 of 638, across 3 files
- Verify each evaluation answer independentlyin 45 of 638, across 3 files
- Implement API client with authentication and paginationin 45 of 638, across 3 files
- Define input schemas with validationin 27 of 638, across 9 files
Said here and by no other author read
- Enforce allow-lists with no default access
- Monitor server health continuously
- Classify all tools by risk level before enabling
- Check dependents before disabling servers
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.