Batch analyze ad creatives (images and videos) using Gemini AI. Returns visual descriptions, text overlays, emotional tone, messaging themes, and video transcripts. Pass URLs from meta-search-ads or google-search-ads results. Images: ~2-3s each (5 concurrent). Videos: ~10-20s each (2 concurrent)....
AI agents call analyze-creatives to retrieve information from Ad Library MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool queries and analyzes existing ad data from public sources (Meta, Google, Instagram, TikTok, YouTube ads) using Gemini AI. It retrieves metadata and creative insights but does not create, modify, delete, or execute any operations with side effects. The caching behavior is internal session state, not a side effect on external systems. This is a classic Read operation with no destructive or write capability.
From the tool's definition Tool description states it 'analyze[s] ad creatives' and 'Returns visual descriptions, text overlays, emotional tone, messaging themes, and video transcripts.' This is purely analytical and retrieval-based—it processes publicly available ad content and…
Attacks that exploit this kind of access
Batch analyze ad creatives (images and videos) using Gemini AI. Returns visual descriptions, text overlays, emotional tone, messaging themes, and video transcripts. Pass URLs from meta-search-ads or google-search-ads results. Images: ~2-3s each (5 concurrent). Videos: ~10-20s each (2 concurrent). Results are cached for the session. It is categorised as a Read tool in the Ad Library MCP MCP Server, which means it retrieves data without modifying state.
Register the Ad Library MCP server in PolicyLayer and add a rule for analyze-creatives: 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 Ad Library MCP. Nothing to install.
analyze-creatives 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 analyze-creatives 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 analyze-creatives. 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.
analyze-creatives is provided by the Ad Library MCP server (thedavidquan01/ad-library-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 →