AI agents use session_speichern to create or update resources in Lokyy — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Lokyy environment.
This tool creates a new record (session source) in the knowledge base by saving chat insights to RAW/sessions/. It is a reversible write operation (data creation/storage), not destructive, financial, or executable. Severity is medium because an AI agent could write unintended or sensitive data into the knowledge base, but the blast radius is limited to the session store.
From the tool's definition 'Hält die Kernerkenntnisse eines Chats als Quelle in RAW/sessions/ fest' — stores/persists core insights of a chat session as a source entry
Attacks that exploit this kind of access
Hält die Kernerkenntnisse eines Chats als Quelle in RAW/sessions/ fest (. It is categorised as a Write tool in the Lokyy MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Lokyy MCP server in PolicyLayer and add a rule for session_speichern: 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 Lokyy. Nothing to install.
session_speichern 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 session_speichern 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 session_speichern. 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.
session_speichern is provided by the Lokyy MCP server (oliverhees/lokyy-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 →