AI agents call codegraph_filter to retrieve information from Context 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 graph data without side effects. It operates on a codebase knowledge graph by filtering nodes based on semantic properties and sorting them—a pure read operation. There is no indication it modifies, deletes, or executes code. It follows the 'Read' category pattern of search/query operations that return information without altering state.
From the tool's definition The tool performs filtering and querying operations on graph nodes with results sorted by PageRank. The description indicates it 'filters' and returns results without modification capabilities.
Attacks that exploit this kind of access
Filter graph nodes by semantic properties. Results sorted by PageRank (most connected first). It is categorised as a Read tool in the Context MCP Server, which means it retrieves data without modifying state.
Register the Context MCP server in PolicyLayer and add a rule for codegraph_filter: 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 Context. Nothing to install.
codegraph_filter 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 codegraph_filter 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 codegraph_filter. 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.
codegraph_filter is provided by the Context MCP server (vibhasdutta/context-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 →