AI agents call local_search to retrieve information from CodeGraphMCPServer without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs localized search queries within a codebase graph structure. It retrieves and queries data about entity relationships and neighborhoods in the code graph, with no side effects, no code execution, and no data modification. This is a classic Read operation typical of code analysis and discovery tools.
From the tool's definition Tool name 'local_search' and description 'Search within entity neighborhood using GraphRAG' indicate a search/query operation.
Documented attack patterns abuse exactly the kind of access local_search gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and CodeGraphMCPServer, and nothing reaches the server without passing your rules. This is the rule we recommend for local_search:
{
"version": "1",
"default": "deny",
"tools": {
"local_search": {}
}
}local_search 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 within entity neighborhood using GraphRAG. It is categorised as a Read tool in the CodeGraphMCPServer MCP Server, which means it retrieves data without modifying state.
Register the CodeGraphMCPServer MCP server in PolicyLayer and add a rule for local_search: 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 CodeGraphMCPServer. Nothing to install.
local_search 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 local_search 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 local_search. 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.
local_search is provided by the CodeGraphMCPServer MCP server (nahisaho/codegraphmcpserver). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from CodeGraphMCPServer, 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.
14 CodeGraphMCPServer tools catalogued and risk-classified — across an index of 43,000+ MCP servers.