tool_compare
AI agents call tool_compare to retrieve information from Codelens-MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Although the description is empty (lowering confidence), the tool name 'compare' and its position among read-only analysis tools (search, list, get, audit, check) strongly suggests this retrieves and compares data without modification. Comparison operations on public repositories have no side effects. The blast radius of misuse is minimal since it operates on public GitHub data.
From the tool's definition Tool name is 'tool_compare' with empty description. Context shows this is a Codelens-MCP server for analyzing public GitHub repositories. Sibling tools like 'tool_search_code', 'tool_get_file', and 'tool_get_package_info' are all read-only operations.
Attacks that exploit this kind of access
tool_compare. It is categorised as a Read tool in the Codelens-MCP MCP Server, which means it retrieves data without modifying state.
Register the Codelens- MCP server in PolicyLayer and add a rule for tool_compare: 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 Codelens-MCP. Nothing to install.
tool_compare 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 tool_compare 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 tool_compare. 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.
tool_compare is provided by the Codelens- MCP server (yashkashte5/codelens-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 →