Mcp

Guardrails for LLM output: pass / fail / review verdicts with calibrated confidence in under 500 ms.

LLM Mart 0 views 4 listing impressions
Transport
Not stated
Package
—
Registry id
ai.overwing/mcp

No install snippet on purpose. A working MCP config is a command, its arguments and an environment block — the last two are where API keys live, so this catalogue never stores them and cannot publish them. Follow the link above for the authors' own instructions.

Overwing returns pass, fail, or review verdicts for each rule with calibrated confidence in under half a second. It offers a free tier with 250 evaluations per day. It integrates with the Vercel AI SDK, the OpenAI Agents SDK, and LangChain.

Pricing is per organization, monthly, and cancelable any time. Every rule is judged in parallel by TypeSafe's Jev model, returning typed answers and probabilities. Every plan includes custom rule sets, webhooks, and a dashboard.

Summary drafted from the project's own website. Every sentence is backed by text on that page and was reviewed before publishing.

Related servers

Fetch Official

Fetch a web page and hand the agent readable text instead of raw HTML.

stdio mcp-server-fetch
27 views
Filesystem Official

Give an agent scoped read and write access to directories you choose.

stdio @modelcontextprotocol/server-filesystem
24 views
Git Official

Read repository history, diffs and branches as tools an agent can call.

stdio mcp-server-git
23 views

Read-only MCP server for the OrchestKit docs: full-text search + Markdown fetch. No auth.

21 views