AI agents use update_prompt to create or update resources in Peecai — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Peecai environment.
This tool modifies metadata (topic and tag assignments) associated with prompts. The change is reversible (assignments can be updated again), poses no irreversible data loss, does not execute code or external operations, and does not involve financial transactions. It clearly fits the Write category (create/modify data reversibly) rather than more severe categories.
From the tool's definition Tool name is 'update_prompt' and description states it can 'Update prompt topic or tag assignments' — a modification operation that reversibly changes data associations without deleting or executing code.
Attacks that exploit this kind of access
Update prompt topic or tag assignments. It is categorised as a Write tool in the Peecai MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Peecai MCP server in PolicyLayer and add a rule for update_prompt: 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 Peecai. Nothing to install.
update_prompt 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 update_prompt 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 update_prompt. 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.
update_prompt is provided by the Peecai MCP server (mcp-server-peecai). 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 →