AI agents call find_implementations to retrieve information from Orihime without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs static code analysis to retrieve implementation relationships from an indexed knowledge graph. It queries existing data (class-interface relationships) without creating, modifying, deleting, or executing any operations. The traversal of up to 10 hops is a read-only graph navigation. No code execution, data mutation, or external side effects are possible.
From the tool's definition Tool description states it 'finds all classes that directly implement the given interface', which is a retrieval/query operation over a code knowledge graph with no modification or side effects.
Attacks that exploit this kind of access
Find all classes that directly implement the given interface (up to 10 hops via IMPLEMENTS). It is categorised as a Read tool in the Orihime MCP Server, which means it retrieves data without modifying state.
Register the Orihime MCP server in PolicyLayer and add a rule for find_implementations: 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 Orihime. Nothing to install.
find_implementations 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 find_implementations 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 find_implementations. 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.
find_implementations is provided by the Orihime MCP server (srinivasan-sundaresan95/orihime). 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 →