AI-citation (GEO) teardowns of any website: score, blockers, attack points.
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
Fetch the summary of a BriefBash teardown by brief_id. While status is "processing", wait check_back_seconds and call again. When done, returns the AI-citation score, potential score, blockers and top attack points. Full interactive teardown is at brief_url.
The brief_id returned by submit_analysis (starts with bb_)
This tool isn't marked read-only, so it may change something on the server or in an account you connected.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http mcp https://briefbash.com/mcp
Fetch the summary of a BriefBash teardown by brief_id. While status is "processing", wait check_back_seconds and call again. When done, returns the AI-citation score, potential score, blockers and top attack points. Full interactive teardown is at brief_url.
| Argument | Type | Description |
|---|---|---|
| brief_id* | string | The brief_id returned by submit_analysis (starts with bb_) |
{
"type": "object",
"required": [
"brief_id"
],
"properties": {
"brief_id": {
"type": "string",
"description": "The brief_id returned by submit_analysis (starts with bb_)"
}
}
}Submit a website URL to BriefBash for an AI-citation (GEO) teardown: how visible the site is to ChatGPT/Perplexity/AI Overviews, what blocks citations (llms.txt, bot access, schema), and the attack points a competitor is winning with. Returns a brief_id — poll get_brief until status is done (typically 1-3 minutes). The first brief per email is free.
| Argument | Type | Description |
|---|---|---|
| url* | string | Absolute http(s) URL of the page or site to analyze, e.g. https://example.com/blog/post |
| email* | string | Email the brief is billed/claimed under (first brief free; later briefs use the email's BriefBash credits) |
{
"type": "object",
"required": [
"url",
"email"
],
"properties": {
"url": {
"type": "string",
"description": "Absolute http(s) URL of the page or site to analyze, e.g. https://example.com/blog/post"
},
"email": {
"type": "string",
"description": "Email the brief is billed/claimed under (first brief free; later briefs use the email's BriefBash credits)"
}
}
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 6, 2026.
No readOnlyHint / destructiveHint annotations, so clients can't tell a read from a write without asking.
Fix: Add annotations to each tool, e.g. { readOnlyHint: true } for tools that only read.