Fabric MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| YTDLP_PATH | No | Custom path to the yt-dlp executable | |
| FABRIC_PATH | No | Custom path to the Fabric CLI executable |
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 | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| health_checkA | Check if Fabric, yt-dlp, and other dependencies are installed and configured correctly. Provides installation instructions if needed. |
| run_fabric_patternA | Run any Fabric pattern with custom input. Use this for patterns not covered by specific tools. |
| get_youtube_transcriptB | Fetch transcript from a YouTube video URL |
| list_fabric_patternsA | List all available Fabric patterns |
| analyze_youtube_videoA | Fetch a YouTube transcript and analyze it with a Fabric pattern in one step. This ensures the complete transcript is processed without truncation. Use this instead of calling get_youtube_transcript and then a pattern separately. |
| analyze_fileA | Read a file and analyze it with a Fabric pattern in one step. This ensures the complete file content is processed without truncation. Useful for analyzing documents, code files, logs, etc. |
| analyze_urlA | Fetch content from a URL and analyze it with a Fabric pattern in one step. Works with articles, documentation, blog posts, etc. This ensures complete content is processed without truncation. |
| fabric_extract_wisdomA | Extract key insights, quotes, and wisdom from any content (articles, videos, podcasts) |
| fabric_summarizeB | Create a concise summary of content |
| fabric_analyze_claimsB | Analyze and fact-check claims made in content |
| fabric_create_quizB | Generate quiz questions from content for learning |
| fabric_to_flashcardsC | Convert content into flashcards for studying |
| fabric_analyze_paperC | Analyze academic papers or technical documents |
| fabric_summarize_git_diffA | Summarize git diff output for code reviews |
| fabric_analyze_logsC | Analyze log files for issues and patterns |
| fabric_analyze_incidentC | Analyze security incidents or system failures |
| fabric_create_coding_projectB | Generate project structure and planning from an idea |
| fabric_explain_codeC | Explain code in simple terms |
| fabric_improve_writingC | Improve writing quality and clarity |
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 19 tools
Each tool has a clear primary purpose, but run_fabric_pattern overlaps with the specific fabric_* tools, and analyze_youtube_video duplicates the combination of get_youtube_transcript and a pattern. These overlaps are mostly resolved by explicit guidance, leaving minor ambiguity.
Naming is mixed: some tools use the fabric_ prefix (fabric_create_coding_project, fabric_summarize) while others do not (health_check, analyze_file). Verbs are clear and readable, but the lack of a uniform prefix or verb-object structure across all tools makes the set less predictable.
With 19 tools, the server is slightly heavy. Many specific fabric_* pattern wrappers could be handled by the generic run_fabric_pattern, making the count feel padded. However, the wide range of content analysis tasks justifies the number to some extent.
The generic run_fabric_pattern ensures any Fabric pattern can be executed, covering missing specific patterns. Health check and list_fabric_patterns support the operational lifecycle, and content ingestion via transcript, file, and URL is comprehensive. No obvious dead ends exist.