List active Fly.io machines used for Proctor exams. Returns information about currently running or recently active Fly machines that are being used for exam execution. Returns: - machines: Array of machine objects with id, state, region, and other metadata Use cases: - Monitor active exam executi...
AI agents call get_machines to retrieve information from Langfuse Observability without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This is a pure read/query operation that retrieves metadata about existing infrastructure without modifying, deleting, or triggering any actions. The use cases (monitoring, finding machines, debugging) are all observational. While the tool could inform decisions about infrastructure cleanup, it does not itself execute those actions—separate tools (destroy_machine, cancel_exam) would be needed for that.
From the tool's definition The tool 'get_machines' returns information about active Fly.io machines: 'List active Fly.io machines', 'Returns information about currently running or recently active Fly machines'.
Attacks that exploit this kind of access
List active Fly.io machines used for Proctor exams. Returns information about currently running or recently active Fly machines that are being used for exam execution. Returns: - machines: Array of machine objects with id, state, region, and other metadata Use cases: - Monitor active exam execution infrastructure - Find machines to clean up or cancel - Debug issues with running exams - Check resource utilization Note: - Machines may be in various states (running, stopped, etc.) - Use destroy_machine to remove machines that are no longer needed - Use cancel_exam to stop a running exam on a specific machine. It is categorised as a Read tool in the Langfuse Observability MCP Server, which means it retrieves data without modifying state.
Register the Langfuse Observability MCP server in PolicyLayer and add a rule for get_machines: 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 Langfuse Observability. Nothing to install.
get_machines is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the get_machines 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 get_machines. 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.
get_machines is provided by the Langfuse Observability MCP server (langfuse-observability-mcp-server). 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 →