AI agents call search_codebase_by_file to retrieve information from Omni Rag without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This is a search/query tool that retrieves information from an indexed codebase based on file pattern filters. It has no side effects, cannot modify data, execute code, or delete content. The deprecation notice does not change its fundamental nature as a read operation. Classified as Read with low severity due to its passive, informational nature.
From the tool's definition Tool performs 'Search filtered by file pattern' - a read-only query operation over indexed files. The description explicitly indicates searching/retrieval without modification or execution capabilities.
Attacks that exploit this kind of access
[Deprecated: use 'search_by_file'] Search filtered by file pattern. It is categorised as a Read tool in the Omni Rag MCP Server, which means it retrieves data without modifying state.
Register the Omni Rag MCP server in PolicyLayer and add a rule for search_codebase_by_file: 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 Omni Rag. Nothing to install.
search_codebase_by_file 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_codebase_by_file 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_codebase_by_file. 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_codebase_by_file is provided by the Omni Rag MCP server (suyash2013/codebase-rag-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 →