AI agents use create_spec to create or update resources in Spec Kit — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Spec Kit environment.
This tool creates new data (specifications) within the management system. It is reversible (the created spec can be deleted or modified) and has no destructive capability. While it modifies the system state, the blast radius is limited to adding a new specification document.
From the tool's definition Tool name 'create_spec' and description 'Create a new specification from a template' indicate data creation/modification. The word 'Create' is a classic Write operation that adds a new entity (a specification document) to the system.
Attacks that exploit this kind of access
Create a new specification from a template. It is categorised as a Write tool in the Spec Kit MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Spec Kit MCP server in PolicyLayer and add a rule for create_spec: 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 Spec Kit. Nothing to install.
create_spec 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_spec 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_spec. 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_spec is provided by the Spec Kit MCP server (@fast-kit/spec-kit). 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 →