AI agents call search_specs to retrieve information from Speclib without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs full-text search over technical specifications and returns summaries—a read-only operation with no side effects. The explicit instruction to use a separate tool (get_spec) for full content retrieval confirms search_specs is purely for querying and discovering data. No data is created, modified, deleted, or executed.
From the tool's definition Tool 'search_specs' retrieves spec summaries via search with parsed tags; description explicitly states 'Use get_spec to retrieve full content', indicating this is a retrieval/discovery tool with no modification or destructive capability.
Attacks that exploit this kind of access
Search for specs. Returns summaries with parsed tags. Use get_spec to retrieve full content. It is categorised as a Read tool in the Speclib MCP Server, which means it retrieves data without modifying state.
Register the Speclib MCP server in PolicyLayer and add a rule for search_specs: 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 Speclib. Nothing to install.
search_specs is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the search_specs 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 search_specs. 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.
search_specs is provided by the Speclib MCP server (rang501/speclib-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 →