AI agents use install_skill to create or update resources in Skill Hub — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Skill Hub environment.
Installing code to the local filesystem is a Write operation (creates/modifies files reversibly via the confirmation flow). Severity is high because installing untrusted or malicious code skills could compromise the system, and the tool auto-detects the agent environment which expands potential blast radius.
From the tool's definition Tool description states 'Install a skill to the local filesystem' with a confirmation pattern. This creates/modifies local files. The description explicitly mentions 'confirm=true to install' indicating irreversible file system modification.
Attacks that exploit this kind of access
Install a skill to the local filesystem. First call shows a preview, then call with confirm=true to install. Auto-detects current CLI environment (detected: ${DETECTED_AGENT}). It is categorised as a Write tool in the Skill Hub MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Skill Hub MCP server in PolicyLayer and add a rule for install_skill: 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 Skill Hub. Nothing to install.
install_skill 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 install_skill 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 install_skill. 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.
install_skill is provided by the Skill Hub MCP server (skillhub-club/mcp-server). 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 →