GET /ads/{ad_id}/insights.
AI agents call get_ad_insights to retrieve information from OpenAI Ads MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool retrieves analytics or performance metrics for an ad without modifying any data. However, it accesses advertising account insights which may contain sensitive business information (spend, performance metrics, audience data), warranting medium severity. No side effects or data modification occurs.
From the tool's definition Tool name 'get_ad_insights' and description 'GET /ads/{ad_id}/insights' indicate a retrieval operation using HTTP GET method. The endpoint pattern and sibling tools (all prefixed with 'get_' or 'list_') confirm this is a read operation.
Attacks that exploit this kind of access
GET /ads/{ad_id}/insights. It is categorised as a Read tool in the OpenAI Ads MCP Server MCP Server, which means it retrieves data without modifying state.
Register the OpenAI Ads MCP Server MCP server in PolicyLayer and add a rule for get_ad_insights: 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 OpenAI Ads MCP Server. Nothing to install.
get_ad_insights 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_ad_insights 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_ad_insights. 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_ad_insights is provided by the OpenAI Ads MCP Server MCP server (roast-labs/openai-ads-mcp). 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 →