Search and read the transcripts of Starter Story's YouTube videos. No account or key needed.
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.
List the videos this guruling learned, newest first: date, length, title, video id and source. Up to 200 at a time; pass offset to go on, and source (a name or id) to keep to one source.
How many, up to 200 (default 100)
How many to skip (default 0)
Keep to one source, by its name or id
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http starter-story https://guruling.com/mcp/starter-story
List the videos this guruling learned, newest first: date, length, title, video id and source. Up to 200 at a time; pass offset to go on, and source (a name or id) to keep to one source.
| Argument | Type | Description |
|---|---|---|
| limit | integer | How many, up to 200 (default 100) |
| offset | integer | How many to skip (default 0) |
| source | string | Keep to one source, by its name or id |
{
"type": "object",
"$schema": "https://json-schema.org/draft/2020-12/schema",
"properties": {
"limit": {
"type": "integer",
"maximum": 200,
"minimum": 1,
"description": "How many, up to 200 (default 100)"
},
"offset": {
"type": "integer",
"maximum": 9007199254740991,
"minimum": 0,
"description": "How many to skip (default 0)"
},
"source": {
"type": "string",
"maxLength": 1000,
"description": "Keep to one source, by its name or id"
}
}
}Read one learned video's transcript word for word, with its title, date, source and link. A long one comes in pages: pass the offset the page before gives.
| Argument | Type | Description |
|---|---|---|
| video* | string | The video id, as list_videos and search_transcripts give it |
| offset | integer | Where to start, from the page before (default 0) |
{
"type": "object",
"$schema": "https://json-schema.org/draft/2020-12/schema",
"required": [
"video"
],
"properties": {
"video": {
"type": "string",
"maxLength": 64,
"minLength": 1,
"description": "The video id, as list_videos and search_transcripts give it"
},
"offset": {
"type": "integer",
"maximum": 9007199254740991,
"minimum": 0,
"description": "Where to start, from the page before (default 0)"
}
}
}Search every transcript this guruling learned for words: all of them must appear, case ignored, each at the start of a word (so "pric" finds "pricing"; a word of three letters or fewer, like "AI", only whole); put a phrase in double quotes to keep it whole. Answers the videos with the most mentions first, each with its date, video id and short snippets. It matches words, not meaning: when little comes back, search again with other words for the same thing (a synonym, the singular, fewer words).
| Argument | Type | Description |
|---|---|---|
| limit | integer | How many videos, up to 20 (default 10) |
| query* | string | Words to find, e.g. pricing "cold email" |
{
"type": "object",
"$schema": "https://json-schema.org/draft/2020-12/schema",
"required": [
"query"
],
"properties": {
"limit": {
"type": "integer",
"maximum": 20,
"minimum": 1,
"description": "How many videos, up to 20 (default 10)"
},
"query": {
"type": "string",
"maxLength": 200,
"minLength": 1,
"description": "Words to find, e.g. pricing \"cold email\""
}
}
}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 16, 2026.