AI agents use add_start_node to create or update resources in Dify MCP — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Dify MCP environment.
The tool creates/adds a node to a Dify workflow definition, modifying the workflow structure reversibly. This is a Write operation (creates/modifies data). Severity is medium because misuse could corrupt workflow definitions or create invalid pipelines, but the impact is confined to workflow configuration and is reversible.
From the tool's definition Tool name is 'add_start_node' which indicates creation of a workflow node; sibling tools include 'add_answer_node', 'add_edge', 'add_end_node', 'add_llm_node', 'add_template_transform_node', and 'create_workflow' — all Write operations that modify workflow…
Attacks that exploit this kind of access
add_start_node. It is categorised as a Write tool in the Dify MCP MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Dify MCP server in PolicyLayer and add a rule for add_start_node: 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 Dify MCP. Nothing to install.
add_start_node 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 add_start_node 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 add_start_node. 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.
add_start_node is provided by the Dify MCP server (taiki-kuraishi/dify-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 →