AI agents call file_recall to retrieve information from Tages without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool reads and retrieves historical context/memories about codebase files from persistent storage. It performs a lookup operation with no side effects on the underlying data. While it returns sensitive context that could influence AI decisions (medium severity if misused to extract architecture secrets), it is purely a Read operation.
From the tool's definition Retrieves memories from SQLite storage matching file path criteria. Uses search/query operations ('Find memories', 'matches by', 'prioritises') with no modification or deletion of stored data.
Attacks that exploit this kind of access
Find memories related to specific file paths — matches by exact path, directory prefix, or reverse prefix. Prioritises anti-pattern and convention memories. It is categorised as a Read tool in the Tages MCP Server, which means it retrieves data without modifying state.
Register the Tages MCP server in PolicyLayer and add a rule for file_recall: 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 Tages. Nothing to install.
file_recall 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 file_recall 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 file_recall. 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.
file_recall is provided by the Tages MCP server (ryantlee25-droid/tages). 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 →