recall_recent
AI agents call recall_recent to retrieve information from Memanto MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves recently stored memories without modifying or deleting data. It follows the pattern of other recall tools on the server that fetch historical data. No side effects or state changes are indicated by the name or server context. Classification as Read is appropriate with high confidence despite empty description, given the consistent pattern of recall-based tools.
From the tool's definition Tool name 'recall_recent' indicates data retrieval. Sibling tools include 'recall', 'recall_as_of', and 'recall_changed_since' which are clearly Read operations for querying memory.
Attacks that exploit this kind of access
recall_recent. It is categorised as a Read tool in the Memanto MCP MCP Server, which means it retrieves data without modifying state.
Register the Memanto MCP server in PolicyLayer and add a rule for recall_recent: 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 Memanto MCP. Nothing to install.
recall_recent 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 recall_recent 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 recall_recent. 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.
recall_recent is provided by the Memanto MCP server (moorcheh-ai/memanto). 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 →