mcp-content-pipeline
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| MCP_CP_X_TOPICS | No | Comma-separated topics (default: AI,tech) | |
| MCP_CP_X_ACCOUNTS | No | Comma-separated X usernames (required for X digest) | |
| MCP_CP_GITHUB_REPO | No | Target repo in owner/repo format (required for sync) | |
| MCP_CP_CLAUDE_MODEL | No | Claude model to use (default: claude-sonnet-4-6) | |
| MCP_CP_GEMINI_MODEL | No | Gemini model for images (default: gemini-3.1-flash-image-preview) | |
| MCP_CP_GITHUB_TOKEN | No | GitHub personal access token (required for sync) | |
| MCP_CP_GITHUB_BRANCH | No | Branch to push to (default: main) | |
| MCP_CP_GEMINI_API_KEY | No | Google AI Studio API key for image generation (required for image generation) | |
| MCP_CP_X_BEARER_TOKEN | No | X API v2 bearer token (required for X digest) | |
| MCP_CP_YOUTUBE_API_KEY | No | YouTube Data API v3 key (only for list_channel_videos) | |
| MCP_CP_IMAGE_OUTPUT_DIR | No | Directory for generated images (default: ~/Downloads) | |
| MCP_CP_SUPADATA_API_KEY | No | Supadata API key for YouTube transcript extraction (required for YouTube analysis) | |
| MCP_CP_ANTHROPIC_API_KEY | Yes | Anthropic API key for Claude analysis | |
| MCP_CP_GITHUB_OUTPUT_DIR | No | Output directory for YouTube analyses (default: content/youtube) | |
| MCP_CP_GITHUB_X_OUTPUT_DIR | No | Output directory for X digests (default: content/x-digest) | |
| MCP_CP_MAX_TRANSCRIPT_TOKENS | No | Max transcript length in tokens (default: 100000) |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| analyse_videoA | Analyse a single YouTube video — extracts the English transcript, generates a TL;DR and key points. Use this tool when the user provides a YouTube URL or asks to analyse a YouTube video. Supports youtube.com/watch, youtu.be, youtube.com/shorts, and youtube.com/live URLs. The video must have English captions. Args: url: YouTube video URL custom_prompt: Additional analysis instructions (optional) |
| analyse_x_feedA | Analyse recent X (Twitter) posts and tweets from specified accounts or configured defaults. Fetches posts from X/Twitter user timelines, filters by topic, and generates a digest with a TL;DR and key points. Use this tool when the user asks about X posts, tweets, Twitter feed, what someone posted on X, or wants a digest of X/Twitter activity. Args: usernames: X/Twitter usernames to analyse (defaults to configured MCP_CP_X_ACCOUNTS) topics: Topics to focus on (defaults to configured MCP_CP_X_TOPICS) hours_back: How far back to fetch posts (default: 24, use 168 for weekly) |
| sync_to_githubA | Push analysed content as markdown files to a GitHub repository. Syncs video analyses and X feed digests to a configured GitHub repo. Use after analyse_video or analyse_x_feed to persist results. Args: analyses: List of analysis result objects from analyse_video commit_message: Git commit message (default: 'Add video analyses') x_digests: Optional list of X digest analysis objects from analyse_x_feed |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 3 tools
The three tools have distinct roles: video analysis, X/Twitter feed analysis, and GitHub syncing. The two 'analyse_' tools could be momentarily confused if a user says 'analyse this content' without naming the source, but their descriptions clearly separate YouTube from X/Twitter.
All tools follow a consistent verb_noun pattern: analyse_video, analyse_x_feed, sync_to_github. The use of 'analyse' for the two source-specific tools and 'sync_to' for the persistence step is predictable and readable.
Three tools are well-scoped for a focused content-analysis-to-GitHub pipeline. Each tool earns its place: two analysis sources and one persistence step, with no redundant or trivial tools.
The surface covers the core pipeline: analyse a video, analyse an X feed, and sync results to GitHub. Minor gaps remain around retrieving or listing previously synced analyses and managing configuration, but agents can work around these for the stated purpose.