AI agents use assert_generate to create or update resources in Assert — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Assert environment.
The tool creates new data artifacts (test scenarios) and can persist them to storage via the optional save functionality. While generation alone would be Read, the explicit mention of optional saving elevates this to Write.
From the tool's definition Tool generates and optionally saves E2E test scenarios ('Optionally save it to Assert for later execution'). This creates new test artifacts that are stored, modifying the test suite state.
Attacks that exploit this kind of access
Generate a ready-to-run E2E test scenario in Assert Markdown format from a plain-English description. Optionally save it to Assert for later execution. It is categorised as a Write tool in the Assert MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Assert MCP server in PolicyLayer and add a rule for assert_generate: 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 Assert. Nothing to install.
assert_generate 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 assert_generate 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 assert_generate. 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.
assert_generate is provided by the Assert MCP server (pixel-funnel/assert-click-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 →