AI agents call aleph_search to retrieve information from Barracuda without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves or queries data from an OpenAleph instance without modifying, deleting, or executing operations. The GET method and search functionality confirm it has no side effects. Severity is low as searching public or authorized data poses minimal risk even with agent misuse.
From the tool's definition Tool performs 'Search entities and documents' via GET request; described as 'searching' with 'natural language queries'; uses GET HTTP method which is idempotent and read-only.
Attacks that exploit this kind of access
Search entities and documents in an OpenAleph instance via GET /api/2/search. It is categorised as a Read tool in the Barracuda MCP Server, which means it retrieves data without modifying state.
Register the Barracuda MCP server in PolicyLayer and add a rule for aleph_search: 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 Barracuda. Nothing to install.
aleph_search 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 aleph_search 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 aleph_search. 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.
aleph_search is provided by the Barracuda MCP server (sourcery-info/barracuda-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 →