Skip to main content
Aescut MCP

What Does The Aescut MCP Server Do?

Aescut’s current MCP tool surface, what check_risk returns, and how to use the data in practice.

Short answer

The Aescut MCP server is read-only. It does not install arbitrary third-party tools for you. Its current job is to help an agent look up entries, search the registry, summarize categories, and compute a conservative risk recommendation before the user proceeds.

That is a deliberate boundary: Aescut is the trust and policy layer first, not a blind package manager.

Current tool surface

  • lookup_skill: fetch a single entry by slug with maintainer, status, risk, permissions, and metadata.
  • search_skills: search the registry with filtering across reviewed and deprecated entries.
  • check_risk: turn review status, staleness, permissions, and intended action into a recommendation such as allow, review, avoid, or block.
  • list_categories: enumerate categories and counts.

What check_risk actually returns

The output is not a marketing badge. It is a conservative recommendation that reflects things humans care about when the consequence of a mistake is real: pending review, deprecated status, broad permissions, stale audits, and a sensitive intended action such as deployment or credential handling.

That means two entries with the same nominal risk tier can still get different recommendations depending on what the user is trying to do with them.

How to use it well

The best pattern is to install Aescut once and keep a companion prompt or skill that tells the agent to consult the registry before it recommends or enables a third-party server. That gives you a repeatable checkpoint instead of trusting each session to remember the rule ad hoc.

Sources and further reading