AI agents call find_files to retrieve information from Restic without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The find_files tool performs a read-only search operation across backup snapshots. It retrieves information about files matching a pattern but does not create, modify, delete, or execute any operations. The blast radius of misuse is minimal—an AI agent could search for sensitive file paths it shouldn't know about, but cannot alter data or trigger external actions. This clearly falls under the Read category.
From the tool's definition Tool name is 'find_files' and description states it 'Search[es] for files matching a pattern across snapshots.' This is a query/search operation with no modification, deletion, or execution of external processes.
Attacks that exploit this kind of access
Search for files matching a pattern across snapshots. It is categorised as a Read tool in the Restic MCP Server, which means it retrieves data without modifying state.
Register the Restic MCP server in PolicyLayer and add a rule for find_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 Restic. Nothing to install.
find_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 find_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 find_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.
find_files is provided by the Restic MCP server (mohsenil85/restic-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 →