AI agents call search_findings to retrieve information from Claudit without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and queries existing smart contract audit data from Solodit without creating, modifying, deleting, executing code, or moving funds. It is designed for information discovery and filtering only, making it a Read category tool with low severity since misuse would only surface existing information rather than cause harmful side effects.
From the tool's definition Tool is described as enabling 'searching over' audit findings and the description emphasizes 'filters for severity, firm, tags' - these are query and retrieval operations with no modification of data. The verb 'search' is explicitly read-only.
Documented attack patterns abuse exactly the kind of access search_findings gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Claudit, and nothing reaches the server without passing your rules. This is the rule we recommend for search_findings:
{
"version": "1",
"default": "deny",
"tools": {
"search_findings": {}
}
}search_findings is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Search Solodit. It is categorised as a Read tool in the Claudit MCP Server, which means it retrieves data without modifying state.
Register the Claudit MCP server in PolicyLayer and add a rule for search_findings: 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 Claudit. Nothing to install.
search_findings 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_findings 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_findings. 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_findings is provided by the Claudit MCP server (marchev/claudit). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Claudit, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
4 Claudit tools catalogued and risk-classified — across an index of 43,000+ MCP servers.