AI agents use update_plan to create or update resources in Tpc — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Tpc environment.
This tool modifies existing plan records. Update operations are categorized as Write because they create or modify data reversibly. The severity is medium rather than high because the blast radius depends on what plans control—if they govern critical system behavior or high-stakes decisions, severity could be higher, but without that context, medium reflects typical update operation risk.
From the tool's definition Tool name 'update_plan' and description 'Update an existing plan' indicate modification of existing data. The operation is reversible (can be updated again) and does not delete or destroy data.
Attacks that exploit this kind of access
Update an existing plan. It is categorised as a Write tool in the Tpc MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Tpc MCP server in PolicyLayer and add a rule for update_plan: 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 Tpc. Nothing to install.
update_plan 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_plan 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_plan. 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_plan is provided by the Tpc MCP server (suttonwilliamd/tpc-server). 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 →