metamcp

metamcp is a skill for Claude Code from VKirill/claude-lane-stack. It costs 29 tokens per session (729 once invoked), scanned A, original, MIT.

A gateway that gives an agent one way to find and call tools from several Model Context Protocol servers. MCP is a standard way for AI agents to use external tools and services.

In plain words
What is it for?
Use it to discover tools, connect a capability to its server, call tools through MetaMCP, and run multi-step JavaScript operations across supported servers.
Why use it?
It keeps some tools directly available while routing others through one discovery and calling layer, instead of requiring separate handling for every server.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: mentions Codex.

Not installable: its command points at a path on the author’s own machine, so it runs nowhere else. The line is /home/ubuntu/.agents/bin/metamcp-run.

Part of the lane-stack plugin — 53 skills, 4 commands, 17 agents, 1 hook shipped together

Good fit Use it to discover tools, connect a capability to its server, call tools through MetaMCP, and run multi-step JavaScript operations across supported servers.

Compare 6 skills from other repositories ↓
Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add VKirill/claude-lane-stack
Claude Code
/plugin install lane-stack

Made for: Claude Code.

Or install lane-stack, the plugin that ships this one along with the rest of its 53 skills, 4 commands, 17 agents, 1 hook.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for metamcp

README.md
[![agentmods](https://agentmods.dev/badge/skills/vkirill/claude-lane-stack/metamcp/github.svg)](https://agentmods.dev/skills/vkirill/claude-lane-stack/metamcp)
Your own site
<a href="https://agentmods.dev/skills/vkirill/claude-lane-stack/metamcp"><img src="https://agentmods.dev/badge/skills/vkirill/claude-lane-stack/metamcp/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for metamcp

Your own site · 80×15
<a href="https://agentmods.dev/skills/vkirill/claude-lane-stack/metamcp"><img src="https://agentmods.dev/badge/skills/vkirill/claude-lane-stack/metamcp.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 29 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 729 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00029 $0.00729
Opus 5 $0.00015 $0.00365
Sonnet 5 $0.00006 $0.00146
Haiku 4.5 $0.00003 $0.00073

Measured 9d ago against content hash fa373dc0d0e2, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

metamcp scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/lane-stack/skills/metamcp/SKILL.md · 89 lines

How it starts

The opening of the file, as written. The whole thing — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.

MetaMCP (hybrid)

Layout

Host-level (direct) Behind MetaMCP (~/.agents/metamcp.mcp.json)
agentmemory context7, postgres, github, shadcn
gitnexus open-design, chrome-devtools, vue-docs
metamcp mcp-books, perplexity, image-gen
remotion-documentation, codex-mcp-server

Never behind MetaMCP (removed): orchestrator-mcp, agent-consult, repowise.

Launcher: /home/ubuntu/.agents/bin/metamcp-run
Config: /home/ubuntu/.agents/metamcp.mcp.json
Env secrets: /home/ubuntu/.agents/metamcp.env (600)

MetaMCP tools (@mentu/[email protected] → 4 tools)

  1. mcp_discover — search tools / list servers (first call can cold-start many children — prefer mcp_call when you know the server)
  2. mcp_provision — resolve capability → server
  3. mcp_call — call a child tool (prefer this)
  4. mcp_execute — multi-step JS sandbox over servers

0.5.x docs mention skill tools; installed 0.4.1 exposes the four core tools only.

How to call (examples)

Direct (no MetaMCP):

  • mcp__agentmemory__memory_recall / memory_remember / …
  • mcp__gitnexus__impact / query / context / detect_changes

Via MetaMCP:

mcp_discover { "query": "screenshot" }
mcp_call {
  "server": "chrome-devtools",
  "tool": "take_screenshot",
  "args": { ... }
}
mcp_call {
  "server": "github",
  "tool": "<tool_name>",
  "args": { ... }
}
mcp_call {
  "server": "perplexity",
  "tool": "perplexity_search",
  "args": { "query": "..." }
}
mcp_call {
  "server": "context7",
  "tool": "resolve-library-id",  # or docs tool name from discover
  "args": { ... }
}

Rules

  1. Prefer direct gitnexus + agentmemory — do not route them through MetaMCP.
  2. For anything else: mcp_discover once if unsure, then mcp_call.
  3. Do not invent host-level MCP servers; add children only in metamcp.mcp.json.
  4. After editing child config, restart CLI sessions (hot-reload may apply inside MetaMCP only).

Read the full file on GitHub · 89 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 9d ago First seen · 89 lines · 29 tokens per session scan A fa373dc0d0e2

Subscribe to this mod's changes

metamcp is a skill published in the GitHub repository VKirill/claude-lane-stack (115 stars, last pushed today), licensed MIT. It adds 29 tokens to every session and 729 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.