codegen_plan · Pen MCP: Risk & Policy

WHAT IT DOES

What codegen_plan does on Pen

AI agents use codegen_plan to create or update resources in Pen — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Pen environment.

WHY IT NEEDS A POLICY

Why codegen_plan needs a policy

This tool creates and stores a new code generation plan, which is a state mutation. It does not execute code generation itself (that happens in codegen_submit_chunk and codegen_assemble), nor does it delete or move money. The validation logic (duplicate detection, dependency checking) indicates this creates a persistent plan object.

From the tool's definition codegen_plan submits and creates a code generation plan that is stored and referenced by subsequent tools (planId is returned and used by codegen_submit_chunk and codegen_assemble). The tool modifies server state by creating a new plan artifact.

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

codegen_plan Rate-limited

codegen_plan stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.

View as policy code

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "codegen_plan": {
      "limits": [
        {
          "counter": "codegen_plan_rate",
          "window": "minute",
          "max": 30,
          "scope": "grant"
        }
      ]
    }
  }
}

LIMIT THIS TOOL →

Instant setup, no code required.

PolicyLayer is an MCP gateway: every tool call is checked against your rule before it reaches Pen, so you can run agents against it safely.

FAQ

Questions about codegen_plan

What does the codegen_plan tool do? +

Submit a code generation plan (CodePlanFromAI). The server validates the plan (rejects on duplicate chunkIds, empty nodeIds, unknown dependencies, circular deps, missing nodeIds) and returns a topologically sorted executionPlan with warnings for shared nodeIds. The planId returned is used by codegen_submit_chunk and codegen_assemble. It is categorised as a Write tool in the Pen MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.

How do I enforce a policy on codegen_plan? +

Register the Pen MCP server in PolicyLayer and add a rule for codegen_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 Pen. Nothing to install.

What risk level is codegen_plan? +

codegen_plan is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.

Can I rate-limit codegen_plan? +

Yes. Add a rate_limit block to the codegen_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.

How do I block codegen_plan completely? +

Set action: deny in the PolicyLayer policy for codegen_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.

What MCP server provides codegen_plan? +

codegen_plan is provided by the Pen MCP server (@zseven-w/pen-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.