agentleFS
Sign inStart free
Google CloudKnown publisherGoogleCloudPlatform/devrel-demos/agents/declared-subagents-demo/skills/compare-subagent-mcp-visibility/SKILL.md

compare-subagent-mcp-visibility

Skill397 starsChanged 11 days agoApache-2.0View on GitHub

Use when comparing subagent MCP server visibility by launching the four declared MCP inspector subagents in parallel (mcp-inspector-default, mcp-inspector-local-only, mcp-inspector-genkit-only, and mcp-inspector-vault-only).

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    When asked to compare subagent MCP configurations or inspect MCP server visibility across subagents, follow these steps:

    line 10
  2. Very specific:

    Call invoke_subagent with all four subagents in parallel so they execute concurrently with the same prompt:

    line 13
  3. Distinctive:

    Wait for all four subagents to report back via reactive messages.

    line 33

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Subagent MCP Comparison Workflow

When asked to compare subagent MCP configurations or inspect MCP server visibility across subagents, follow these steps:

  1. Launch All 4 Declared Subagents in Parallel:
    • Call invoke_subagent with all four subagents in parallel so they execute concurrently with the same prompt: "Inspect your <mcp_servers> context block and native tool declarations without reading any files. Report: (1) Which MCP servers and tools you see in <mcp_servers>, and (2) Which native tools you have (including call_mcp_tool).":
      1. mcp-inspector-default:
        • TypeName: "mcp-inspector-default"
        • Role: "Default MCP Inspector (No mcpServers)"
        • Workspace: "inherit"
      2. mcp-inspector-local-only:
        • TypeName: "mcp-inspector-local-only"
        • Role: "Local-Only MCP Inspector (local-db-server)"
        • Workspace: "inherit"
      3. mcp-inspector-genkit-only:
        • TypeName: "mcp-inspector-genkit-only"
        • Role: "Genkit-Only MCP Inspector (genkit-mcp-server)"
        • Workspace: "inherit"
      4. mcp-inspector-vault-only:
        • TypeName: "mcp-inspector-vault-only"
        • Role: "Vault-Only MCP Inspector (inline secret-vault-server)"
        • Workspace: "inherit"
        • Note: Inline MCP servers are started asynchronously on their first launch in a session. If secret-vault-server has not finished initializing by Turn 1 of mcp-inspector-vault-only, send a follow-up message (send_message) to mcp-inspector-vault-only asking it to re-inspect <mcp_servers> on Turn 2 once initialization completes.

  2. Present the Comparison Table:

    • Wait for all four subagents to report back via reactive messages.
    • Present a single, compact Markdown comparison table (no lengthy prose) using ✅ and ❌ to show server visibility at a glance:
    SubagentmcpServers Configlocal-db-server (Plugin)genkit-mcp-server (Global)secret-vault-server (Inline)Behavior
    mcp-inspector-default(omitted)✅ / ❌✅ / ❌✅ / ❌Short 3–5 word summary (e.g., Inherits global + plugin servers)
    mcp-inspector-local-onlylocal-db-server (by name)✅ / ❌✅ / ❌✅ / ❌Short 3–5 word summary (e.g., Plugin server only; drops global)
    mcp-inspector-genkit-onlygenkit-mcp-server (by name)✅ / ❌✅ / ❌✅ / ❌Short 3–5 word summary (e.g., Plugin + named global server)
    mcp-inspector-vault-onlysecret-vault-server (inline)✅ / ❌✅ / ❌✅ / ❌Short 3–5 word summary (e.g., Plugin + private inline server)

Constraints:

  • Do NOT view, read, or inspect source code files (e.g. mcp_server.py or secret_vault_server.py) or previous transcripts to extract data.
  • Do NOT execute shell commands or script files directly via run_command to query data.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free