end_session
AI agents use end_session to create or update resources in Astria-Index — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Astria-Index environment.
This tool appears to close or finalize a session object within the memory system. Since it modifies session state (completing/archiving it) rather than retrieving data (Read), executing arbitrary code (Execute), or permanently removing records (Destructive), it fits the Write category. It's reversible—sessions can typically be reopened or their data recovered.
From the tool's definition Tool name 'end_session' indicates termination or closure of a session. In context of an AI memory system, ending a session likely commits or finalizes session state, which is a reversible state modification. The description is empty, reducing confidence.
Attacks that exploit this kind of access
end_session. It is categorised as a Write tool in the Astria-Index MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Astria-Index MCP server in PolicyLayer and add a rule for end_session: 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 Astria-Index. Nothing to install.
end_session is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the end_session 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 end_session. 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.
end_session is provided by the Astria-Index MCP server (pl-odin/astria-plugin). 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 →