Search the codebase for architecture concepts, execution flows, and file matches using hybrid vector/AST search. Supply projectId to scope to a specific project.
AI agents call corn_code_search to retrieve information from CornMCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
corn_code_search is a read-only tool that searches and retrieves information from a codebase. It enables queries across architecture concepts, execution flows, and file matches but does not create, modify, delete, or execute any code. The scope parameter allows filtering by project but does not enable destructive or mutative operations.
From the tool's definition Tool performs 'search' and 'matches' operations on codebase using 'hybrid vector/AST search' with no mention of modifications, deletions, or side effects. Description explicitly indicates retrieval/query behavior.
Documented attack patterns abuse exactly the kind of access corn_code_search gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and CornMCP, and nothing reaches the server without passing your rules. This is the rule we recommend for corn_code_search:
{
"version": "1",
"default": "deny",
"tools": {
"corn_code_search": {}
}
}corn_code_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 the codebase for architecture concepts, execution flows, and file matches using hybrid vector/AST search. Supply projectId to scope to a specific project. It is categorised as a Read tool in the CornMCP MCP Server, which means it retrieves data without modifying state.
Register the Corn MCP server in PolicyLayer and add a rule for corn_code_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 CornMCP. Nothing to install.
corn_code_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 corn_code_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 corn_code_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.
corn_code_search is provided by the Corn MCP server (yuki-20/cornmcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 18 CornMCP tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
18 CornMCP tools catalogued and risk-classified — across an index of 42,500+ MCP servers.