Start an OpenTelemetry trace for an agent turn. Call at the beginning of handling each user query.
AI agents use start_turn to create or update resources in Cursor Otel — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Cursor Otel environment.
This tool creates a new OpenTelemetry trace entry at the start of each agent turn. It writes telemetry/observability data but has no destructive, financial, or execution side effects. Misuse would result in spurious trace records at worst, making severity low.
From the tool's definition 'Start an OpenTelemetry trace for an agent turn' — creates a new telemetry trace/span record
Attacks that exploit this kind of access
Start an OpenTelemetry trace for an agent turn. Call at the beginning of handling each user query. It is categorised as a Write tool in the Cursor Otel MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Cursor Otel MCP server in PolicyLayer and add a rule for start_turn: 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 Cursor Otel. Nothing to install.
start_turn 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 start_turn 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 start_turn. 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.
start_turn is provided by the Cursor Otel MCP server (smith/cursor-otel). 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 →