Skill Detail

Share MCP tools and agent skills across local agents with mcptoon

Use mcptoon to install, index, inspect, and call MCP tools and agent skills from one local CLI while keeping bulky schemas and skill text out of every agent context.

Integrations & ConnectorsMulti-Framework
Integrations & Connectors Multi-Framework Security Reviewed
⭐ 207 GitHub stars
COPY SKILL INSTRUCTIONS (OPTIONAL)
npx skills add agentskillexchange/skills --skill share-mcp-tools-and-agent-skills-across-local-agents-with-mcptoon Copy
Uses the third-party skills CLI, not an ASE-owned installer. Check your agent’s compatibility. This copies instructions; complete the upstream tool setup below separately.
At a glance
Tools required
Python 3, pip, local MCP servers or agent skills to index, and one or more supported local agents
Install & setup
pip install mcptoon && mcptoon quickstart
Author
activeing123
Publisher
Open Source Project
Last updated
Oct 4, 2026
Quick brief

Use mcptoon when an operator needs one local control plane for MCP servers and agent skills across Claude Code, Codex, Cursor, Windsurf, Cline, VS Code Copilot, and other local agents. The upstream project documents tool discovery, skill search, compact manifests, on-demand schema inspection, direct tool calls, MCP serving, TOON output, starter packs, and quickstart registration for detected agents.

How it works

What this skill actually does

Invoke this instead of hand-editing each agent’s MCP and skill configuration when the job is to share tools across runtimes, reduce context spent on tool schemas, or let an agent discover and inspect tools on demand. The scope boundary is local MCP and skill management plus context compaction. It is not a general package registry, model framework, or standalone application platform.

Inputs and prerequisites: Python 3, pip, local MCP servers or agent skills to index, and one or more supported local agents.

Setup notes: pip install mcptoon && mcptoon quickstart

Source and verification boundary: use https://activeing123.github.io/mcptoon/ as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.

Framework fit: publish this as a Multi-Framework workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.