Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/rune-kit/rune/mcp-buildergit clone --depth 1 https://github.com/Rune-kit/runeWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00039 | $0.00272 |
| Opus 5 | $0.00019 | $0.00136 |
| Sonnet 5 | $0.00008 | $0.00054 |
| Haiku 4.5 | $0.00004 | $0.00027 |
Grade A, and why
mcp-builder 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 2d 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.
What it actually says
You are the mcp-builder skill — Rune's MCP server generator.
Quick Reference
Workflow:
- Spec — parse user description or specification for tools/resources to expose
- Research — look up target API docs, existing MCP servers for reference
- Generate — tool definitions, resource handlers, input validation, error handling, config
- Test — generate and run test suite for the server
- Docs — generate server documentation (tool catalog, installation, configuration)
- Verify — build passes, tests pass, server starts
Supports: TypeScript (official @modelcontextprotocol/sdk) and Python (FastMCP)
Dual-Model Config: Use haiku for simple tool wrappers, sonnet for complex multi-resource servers.
Called by: cook (MCP task detected), scaffold (MCP Server template). Manual: /rune mcp-builder.
Read skills/mcp-builder/SKILL.md for the full specification including adapter patterns.
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.
- 2d ago First seen · 27 lines · 39 tokens per session scan A 0a150d5b032a
mcp-builder is an agent published in the GitHub repository Rune-kit/rune (84 stars, last pushed 17d ago), licensed MIT. It adds 39 tokens to every session and 272 once invoked, about $0.0002 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.
Other agents, from other repositories
polymath
Cross-disciplinary synthesis; spawns domain-specific subagents, synthesizes findings across domains, and produces integrated insights.
probe-real-agent
The only file in this directory that is meant to be a subagent. If the runtime registers exactly this one, the probe's recorded behaviour has CHANGED.
codemap
Defines agent personalities (Orchestrator, Explorer, Librarian, etc.) and manages their configuration lifecycle. This directory implements the Agent Factory Pattern, where each agent is a specialized sub-agent with distinct capabilities, permissions, and routing rules. The Orchestrator agent (src/agents/index.ts)…
structured-outputs
Learn how to use structured outputs with an agent.
looping
Re-invoke agents safely with bounded loops, completion evaluators, AI judges, progress feedback, and approval escape behavior.
command-pack
不要把产品白名单写进 core/runtime，也不要通过扫描全局 nodemodules 自动发现包。通用机制放 runtime，产品差异只由 policy 表达。.