Declare what an agent is good at, so the PM can auto-assign matching tasks.
AI agents use set_skills to create or update resources in Claude Team MCP — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Claude Team MCP environment.
The tool modifies agent attributes (skills/capabilities) in the coordination system, which is a write operation. The severity is low because misuse would only affect task assignment logic without direct impact on code, data, or external systems. This is administrative metadata management that can be corrected or undone by declaring different skills.
From the tool's definition Tool description states 'Declare what an agent is good at', indicating it creates or modifies agent skill metadata. The action is reversible and has no destructive, financial, or code execution implications.
Attacks that exploit this kind of access
Declare what an agent is good at, so the PM can auto-assign matching tasks. It is categorised as a Write tool in the Claude Team MCP MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Claude Team MCP server in PolicyLayer and add a rule for set_skills: 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 Claude Team MCP. Nothing to install.
set_skills 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 set_skills 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 set_skills. 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.
set_skills is provided by the Claude Team MCP server (lakshan12367/mcp). 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 →