All skills
PR, Reputation & Crisis
AI-SERP Reputation Audit
See — and source-trace — exactly what Google's AI tells the public about your brand.
Run this skill with your agent
1. Connect the MCP server
https://apidirect.io/mcp?token=YOUR_API_KEY
2. Then just say to your agent
What does Google's AI say about {brand}, and which sources is it pulling that from?
In clients that support MCP prompts (Claude Desktop, Claude Code, Cursor) this skill appears as a prompt named
ai-serp-reputation-audit.
Any agent can also call get_skill(skill_id="ai-serp-reputation-audit") to pull these steps on demand.
What it does
Captures Google's AI Overview and AI Mode answer about your brand, then traces and inspects the cited sources shaping that narrative so you can fix the inputs.
Who it's for: PR, brand and SEO teams managing how AI describes them.
Inputs
| Input | Required | Description | Example |
|---|---|---|---|
brand |
Yes | The brand or person to audit. | API Direct |
How your agent runs it
-
1
search_web(query="{brand} reviews", include_ai_overview=true)Read ai_overview.text_parts (the public summary) and collect ai_overview.reference_links.
-
2
google_ai_mode(prompt="What is the reputation of {brand}? Note complaints and praise.")Get the fuller conversational answer plus its reference_links citations.
-
3
search_web(query="site:<cited_domain> {brand}")Open each citing domain to see the exact pages driving the AI's narrative.
Delivers: A report: what AI says about the brand, the sources it cites, and the specific pages to fix or pitch.
Tools used
The full playbook
This is exactly what the MCP returns to your agent (via the ai-serp-reputation-audit prompt or get_skill tool), with your inputs filled in.
SKILL: AI-SERP Reputation Audit
Captures Google's AI Overview and AI Mode answer about your brand, then traces and inspects the cited sources shaping that narrative so you can fix the inputs.
You are running this skill on API Direct via its MCP tools. Execute the steps below yourself by calling the named tools in order — values in <angle brackets> come from a previous step. Then deliver the result described at the end.
INPUTS:
- brand (required): <brand — ASK THE USER>
STEPS:
1. Tool `search_web` — search_web(query="<brand> reviews", include_ai_overview=true)
Read ai_overview.text_parts (the public summary) and collect ai_overview.reference_links.
2. Tool `google_ai_mode` — google_ai_mode(prompt="What is the reputation of <brand>? Note complaints and praise.")
Get the fuller conversational answer plus its reference_links citations.
3. Tool `search_web` — search_web(query="site:<cited_domain> <brand>")
Open each citing domain to see the exact pages driving the AI's narrative.
DELIVER: A report: what AI says about the brand, the sources it cites, and the specific pages to fix or pitch.
Note: each underlying tool call is billed at its normal endpoint price; get_sentiment adds a small per-page surcharge. Page through results as needed but stop once you have enough to deliver the outcome.