Create a plan graph (DAG of tasks) for an existing run.
AI agents use create_plan_graph to create or update resources in Dag Planner — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Dag Planner environment.
This tool creates (writes) new structured data—a DAG and its constituent tasks—into the workflow run state. While creation is reversible in principle (tasks can be deleted or a new plan can overwrite the old), it materially alters the workflow's execution plan and can trigger downstream task execution.
From the tool's definition Tool description states it will 'Create a plan graph' — a generative action that modifies the workflow state by adding a directed acyclic graph (DAG) of tasks to an existing run.
Attacks that exploit this kind of access
Create a plan graph (DAG of tasks) for an existing run. It is categorised as a Write tool in the Dag Planner MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Dag Planner MCP server in PolicyLayer and add a rule for create_plan_graph: 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 Dag Planner. Nothing to install.
create_plan_graph 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 create_plan_graph 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 create_plan_graph. 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.
create_plan_graph is provided by the Dag Planner MCP server (shubhamnegi/dag-planner-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 →