AI agents use save_session to create or update resources in Strudel — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Strudel environment.
This tool writes/persists session data to storage, making it a Write operation. It is reversible (sessions can be re-saved or deleted by other means) and has no financial impact. Severity is medium because an agent could inadvertently save corrupted or unwanted session states, but the context of a music live-coding environment means the blast radius is limited to that session's data, not system-wide resources.
From the tool's definition Tool name 'save_session' combined with description 'Save a session to storage' indicates a write operation that persists data to storage. The tool creates or modifies stored session state.
Attacks that exploit this kind of access
Save a session to storage. It is categorised as a Write tool in the Strudel MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Strudel MCP server in PolicyLayer and add a rule for save_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 Strudel. Nothing to install.
save_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 save_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 save_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.
save_session is provided by the Strudel MCP server (utenadev/strudel-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 →