Searches YouTube videos via SerpAPI, returning title, link, channel, views, duration, and published date.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@mcp-youtube-search search for python tutorials"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
An MCP server that searches YouTube through SerpApi, for Claude, Cursor and other MCP clients: titles, links, channels, views, publish dates, durations and descriptions.
Python 3.10 or later (or uv)
A SerpApi key in SERP_API_KEY (serpapi.com has a free tier)
Related MCP server: YouTube MCP Server
{
"mcpServers": {
"youtube-search": {
"command": "uvx",
"args": ["mcp-youtube-search", "server"],
"env": { "SERP_API_KEY": "your-serpapi-key" }
}
}
}Or pip install mcp-youtube-search and use "command": "mcp-youtube-search", "args": ["server"].
export SERP_API_KEY="your-serpapi-key"
mcp-youtube-search search "model context protocol" --max-results 5
mcp-youtube-search server # MCP over stdio
mcp-youtube-search server --connection_type http --port 3002 # Streamable HTTP at http://127.0.0.1:3002/mcpLOG_LEVEL=DEBUG shows more detail on stderr. The API key is never logged.
Tool | What it does |
| Videos for a query. |
uv sync
uv run pytest
uv run ruff check .MIT
Search YouTube for videos: title, link, channel, views, publish date, duration and description.
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | What to search for | |
| country | No | Country code, such as us | us |
| language | No | Interface language code, such as en | en |
| max_results | No | How many videos to return (1-50) |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true and openWorldHint=true, so the safety and external-network profile is covered. The description adds only that video metadata (title, link, channel, views, etc.) comes back, which is largely redundant with the output schema; it discloses nothing about quotas, rate limits, or result quality.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
A single front-loaded sentence with no filler, and the verb+resource leads. The trailing field enumeration is partly wasteful given an output schema exists, keeping it just short of a 5.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
With an output schema and annotations carrying the return shape and safety profile, the description only needs to establish purpose, which it does. The gap is the absence of any usage guidance for a tool with no alternatives, which is a minor omission.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100% and every parameter (query, country, language, max_results) is documented in the schema with defaults and bounds. The description adds no parameter-level detail, so the baseline 3 applies.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description names a specific verb (Search) and resource (YouTube videos) and even enumerates the return fields, so the agent knows exactly what the tool produces. There are no sibling tools to differentiate from, so nothing further is required on that front.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Usage is only implied: the agent can infer it should call this to find videos on YouTube, but the description gives no when-to-use framing, no exclusions, and no indication of what to do when the query is too broad or returns nothing.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v0.2.1search_youtubeScored across 1 tool
With only a single tool, there is no possibility of overlap or misselection between tools. The one tool's purpose (YouTube video search) is unambiguous.
The lone tool uses a clean verb_noun snake_case pattern (search_youtube). There is no inconsistent convention to conflict with it.
A single tool is thin even for a search-focused server; it covers exactly one query type. It earns its place but leaves the surface feeling minimal.
Search alone leaves notable gaps: no tool to fetch a specific video's full details, channel info, comments, or transcripts. Core search is covered but agents will hit dead ends for follow-up queries.
YouTube search, whole channels, video stats, comments and full transcripts with timestamps.
YouTube video search with transcript extraction as first-class output.
YouTube transcripts, search, channel/playlist listings and upload tracking for AI agents.
Search YouTube, read video metadata, and fetch transcripts with language preferences
