Agent driven file inspection that uses AI to retrieve relevant content from multiple files.
AI agents call inspect_files to retrieve information from Elastic Brain ZH 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 file content without creating, modifying, or deleting data. It is a Read category operation. Severity is medium rather than low because file inspection in an AI memory system could expose sensitive knowledge stored in Elasticsearch, and the AI-driven nature means the agent's choice of which files to inspect is autonomous.
From the tool's definition Tool description states 'retrieve relevant content from multiple files' which is a read operation. The name 'inspect_files' and verb 'retrieve' indicate data access without modification.
Attacks that exploit this kind of access
Agent driven file inspection that uses AI to retrieve relevant content from multiple files. It is categorised as a Read tool in the Elastic Brain ZH MCP Server, which means it retrieves data without modifying state.
Register the Elastic Brain ZH MCP server in PolicyLayer and add a rule for inspect_files: 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 Elastic Brain ZH. Nothing to install.
inspect_files 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 inspect_files 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 inspect_files. 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.
inspect_files is provided by the Elastic Brain ZH MCP server (tocharianou/elastic-brain-zh). 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 →