AI agents invoke aat_run_skill_mode to trigger actions in AWT (AI Watch Tester). 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 Playwright-based E2E tests against web applications. Running automated browser tests can trigger unintended side effects on the target application—creating test data, modifying state, or interacting with real services. The 'skill_mode' variant suggests autonomous or parameterized test execution, increasing the blast radius if an AI agent misuses it or targets the wrong application.
From the tool's definition Tool name 'aat_run' in sibling tools and 'run' in 'aat_run_skill_mode' indicate execution of test scenarios. Server description states it 'runs Playwright tests' and 'self-heals failures automatically', confirming code execution capabilities.
Documented attack patterns abuse exactly the kind of access aat_run_skill_mode gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and AWT (AI Watch Tester), and nothing reaches the server without passing your rules. This is the rule we recommend for aat_run_skill_mode:
{
"version": "1",
"default": "deny",
"tools": {
"aat_run_skill_mode": {
"limits": [
{
"counter": "aat_run_skill_mode_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}aat_run_skill_mode 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.
aat_run_skill_mode. It is categorised as a Execute tool in the AWT (AI Watch Tester) MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the AWT (AI Watch Tester) MCP server in PolicyLayer and add a rule for aat_run_skill_mode: 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 AWT (AI Watch Tester). Nothing to install.
aat_run_skill_mode 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 aat_run_skill_mode 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 aat_run_skill_mode. 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.
aat_run_skill_mode is provided by the AWT (AI Watch Tester) MCP server (ksgisang/ai-watch-tester). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from AWT (AI Watch Tester), 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.
12 AWT (AI Watch Tester) tools catalogued and risk-classified — across an index of 43,000+ MCP servers.