/ * Triggers the Gemini analysis, waits for completion, and returns the final report. * * This is a synchronous, blocking call that runs the full analysis. * The tool will not return a response until the investigation is complete. * * Workflow: This tool MUST be called immediately after * 'create...
Risk signalsAdmin/system-level operation
Part of the Gemini Cloud Assist server.
Free to start. No card required.
AI agents invoke run_investigation to trigger processes or run actions in Gemini Cloud Assist. Execute operations can have side effects beyond the immediate call -- triggering builds, sending notifications, or starting workflows. Rate limits and argument validation are essential to prevent runaway execution.
run_investigation can trigger processes with real-world consequences. An uncontrolled agent might start dozens of builds, send mass notifications, or kick off expensive compute jobs. PolicyLayer enforces rate limits and validates arguments to keep execution within safe bounds.
Execute tools trigger processes. Rate-limit and validate arguments to prevent unintended side effects.
{
"version": "1",
"default": "deny",
"tools": {
"run_investigation": {
"limits": [
{
"counter": "run_investigation_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}See the full Gemini Cloud Assist policy for all 4 tools.
These attack patterns abuse exactly the kind of access run_investigation gives an agent. Each links to the full case and the policy that stops it:
Other execute tools across the catalogue. The same approach applies to each: rate-limit and validate the arguments.
/ * Triggers the Gemini analysis, waits for completion, and returns the final report. * * This is a synchronous, blocking call that runs the full analysis. * The tool will not return a response until the investigation is complete. * * Workflow: This tool MUST be called immediately after * 'create_investigation' or 'add_observation'. The string returned by this function is the final, * detailed report. There is no need to call any other tool after this to get the * results. * * Interpreting the Output: The output of this tool represents a set of observations & hypotheses based on the data, * not a definitive conclusion. NEVER treat the output as a confirmed root cause without further validation. * Your job is to analyze the returned hypotheses and use other tools at your disposal to gather evidence that proves or disproves them. * * @returns {string} A detailed troubleshooting report in a structured string format. The * report is organized with '##' headers for sections like 'Issue', * 'Hypotheses', 'Relevant Observations', and 'Remediation'. * Note: Some sections may be empty and will state "No hypotheses * found." or similar. Your job is to parse this report and present * a clear summary of the findings (or lack thereof) to the user. */. It is categorised as a Execute tool in the Gemini Cloud Assist MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Gemini Cloud Assist MCP server in PolicyLayer and add a rule for run_investigation: 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 Gemini Cloud Assist. Nothing to install.
run_investigation 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 run_investigation 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 run_investigation. 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.
run_investigation is provided by the Gemini Cloud Assist MCP server (@google-cloud/gemini-cloud-assist-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 4 Gemini Cloud Assist tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
4,600+ MCP servers and 31,000+ tools scanned and risk-classified.