search_pairs
AI agents call search_pairs to retrieve information from Weighted Compact without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This is a read-only search operation over local memory substrate with no side effects, no modifications, and no external operations. Even though the description is empty, the server-level documentation clearly categorizes it as read-only, and the naming pattern 'search_pairs' is consistent with retrieval operations. Blast radius is minimal—worst case returns unexpected search results from local indexed data.
From the tool's definition Tool name 'search_pairs' suggests querying/searching functionality. Server description explicitly states 'Three read-only MCP tools' including this one, and indicates 'zero outbound calls.' The tool operates over local ~/.claude/projects/ data with…
Attacks that exploit this kind of access
search_pairs. It is categorised as a Read tool in the Weighted Compact MCP Server, which means it retrieves data without modifying state.
Register the Weighted Compact MCP server in PolicyLayer and add a rule for search_pairs: 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 Weighted Compact. Nothing to install.
search_pairs 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_pairs 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_pairs. 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_pairs is provided by the Weighted Compact MCP server (zzallirog/weighted-compact). 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 →