Execute code directly in the connected test client and get back the updated DOM state and console logs. IMPORTANT: Before using this tool, call get_current_test_state first to check the
AI agents invoke execute_test_step to trigger actions in Testing. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
This tool executes code in a live test client environment, which is a direct code execution capability. While constrained to a test environment (reducing blast radius vs. production execution), it still allows an AI agent to run arbitrary code with side effects including DOM manipulation, network requests, or other state changes.
From the tool's definition "Execute code directly in the connected test client" - the tool runs arbitrary code in a test environment with access to DOM and console. The description explicitly states it executes code and returns DOM state and logs.
Documented attack patterns abuse exactly the kind of access execute_test_step gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Testing, and nothing reaches the server without passing your rules. This is the rule we recommend for execute_test_step:
{
"version": "1",
"default": "deny",
"tools": {
"execute_test_step": {
"limits": [
{
"counter": "execute_test_step_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}execute_test_step stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Execute code directly in the connected test client and get back the updated DOM state and console logs. IMPORTANT: Before using this tool, call get_current_test_state first to check the. It is categorised as a Execute tool in the Testing MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Testing MCP server in PolicyLayer and add a rule for execute_test_step: 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 Testing. Nothing to install.
execute_test_step is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the execute_test_step 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 execute_test_step. 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.
execute_test_step is provided by the Testing MCP server (mcpland/testing-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Testing, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
5 Testing tools catalogued and risk-classified — across an index of 43,000+ MCP servers.