Find breakout YouTube videos, Reddit/HN topics, hook scores & X/LinkedIn posts. 5 free scans.
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
Bearer vrt_live_... API key from https://verityanalytics.adityaaadi.cv/account
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.
Analyze opening 30-second video hooks to predict audience retention curves (0-100 score) and script optimization recommendations. Counts toward your scan quota (5/year free, 100/day paid resetting at 00:00).
Opening video script lines or hook transcript to audit.
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 verity-analytics https://verityanalytics.adityaaadi.cv/api/mcp \ --header 'Authorization: Bearer YOUR_TOKEN'
Headers
Authorization (required): Bearer vrt_live_... API key from https://verityanalytics.adityaaadi.cv/accountAnalyze opening 30-second video hooks to predict audience retention curves (0-100 score) and script optimization recommendations. Counts toward your scan quota (5/year free, 100/day paid resetting at 00:00).
| Argument | Type | Description |
|---|---|---|
| hook* | string | Opening video script lines or hook transcript to audit. |
{
"type": "object",
"required": [
"hook"
],
"properties": {
"hook": {
"type": "string",
"maxLength": 2500,
"minLength": 10,
"description": "Opening video script lines or hook transcript to audit."
}
},
"additionalProperties": false
}Scan YouTube, Reddit, and Hacker News to compute live content demand gap scores and Views-Per-Hour (VPH) outliers. Free accounts include 5 scans per year within a 30-day window; paid accounts include 100 scans per day resetting at 00:00.
| Argument | Type | Description |
|---|---|---|
| niche* | string | Target content niche or topic query (e.g., "AI coding agents"). |
{
"type": "object",
"required": [
"niche"
],
"properties": {
"niche": {
"type": "string",
"maxLength": 120,
"minLength": 2,
"description": "Target content niche or topic query (e.g., \"AI coding agents\")."
}
},
"additionalProperties": false
}Generate Twitter/X threads, LinkedIn posts, and high-CTR title variations from a validated YouTube outlier. Counts toward your scan quota (5/year free, 100/day paid resetting at 00:00).
| Argument | Type | Description |
|---|---|---|
| title* | string | Title of the validated YouTube outlier video. |
| channel* | string | Creator or channel name of the reference video. |
{
"type": "object",
"required": [
"title",
"channel"
],
"properties": {
"title": {
"type": "string",
"maxLength": 300,
"minLength": 3,
"description": "Title of the validated YouTube outlier video."
},
"channel": {
"type": "string",
"maxLength": 120,
"minLength": 1,
"description": "Creator or channel name of the reference video."
}
},
"additionalProperties": false
}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 22, 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.