AI agents use skillet_install to create or update resources in Skillet — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Skillet environment.
This tool creates new files and modifies the project structure by installing skills. While not destructive (files can theoretically be removed), it is Write-category because it modifies the local project state. Severity is medium because installing untrusted or malicious skills could compromise the project, but the blast radius is limited to the local project scope and the registry is presumably curated.
From the tool's definition Tool name 'skillet_install' performs installation action, description states 'Install a skill from the registry into this project (.claude/skills/)', which creates or modifies project files irreversibly by adding new skill files to the project directory.
Attacks that exploit this kind of access
Install a skill from the registry into this project (.claude/skills/). Pass the registry. It is categorised as a Write tool in the Skillet MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Skillet MCP server in PolicyLayer and add a rule for skillet_install: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Skillet. Nothing to install.
skillet_install is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the skillet_install rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for skillet_install. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
skillet_install is provided by the Skillet MCP server (jnmetacode/skillet). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →