Skip to main content
Glama

jev-av-analysis-mcp

πŸ‡¨πŸ‡³ δΈ­ζ–‡ζ–‡ζ‘£

Local-first, zero-key structured audio/video analysis: download β†’ transcribe β†’ decide in one pipeline. Transcription runs on your own machine with yt-dlp + whisper.cpp; analysis uses TypeSafe Jev (System One decision model) to emit structured labels, not prose.

An MCP server that turns any video/audio into "transcript + structured analysis": topic tags, sentiment tone, sensitivity flags (profanity / violence / medical advice / financial advice / politics / copyrighted music / harassment), and a publish gate (publish / review / block). Built for podcasters, video creators, YouTubers β€” anyone who wants an agent to auto-tag and screen their content.

npm install && npm run build
node dist/index.js doctor      # print transcription deps + Jev mode diagnostics

Connect node dist/index.js in your MCP client and use the four tools below.


The four tools

Tool

Purpose

transcribe_media

URL or local file β†’ transcript (URL via yt-dlp, local/downloaded via whisper.cpp)

fetch_media

Use yt-dlp to pull the best audio track of a video/audio to local; returns path + title + duration

transcribe_file

Local audio/video file β†’ transcript

analyze_transcript

Transcript β†’ Jev structured analysis (see below)

analyze_transcript takes a piece of text (usually from the three transcription tools above) and returns:

{
  "topics":            { "tech": 0.9, "business": 0.1, "education": 0.3, "...": 0.0 },
  "dominant_topics":   ["tech"],
  "sentiment":         { "score": 3, "label": "positive", "confidence": 0.8, "probabilities": {"0":0,"1":0,"2":0.1,"3":0.8,"4":0.1} },
  "sensitivity":       { "profanity": 0.1, "violence": 0.0, "medical_advice": 0.2, "financial_advice": 0.9, "...": 0.0 },
  "flagged_sensitivity": ["financial_advice"],
  "has_key_claims":    0.8,
  "has_engagement_hook": 0.4,
  "publish_gate":      "review",
  "confidence":        0.8,
  "mocked":            false
}

Fields

Field

Meaning

topics / dominant_topics

Membership probabilities over 8 default topics (>0.5 β†’ dominant); override via categories

sentiment

5-level ordinal: very_negative β†’ very_positive, with distribution and confidence

sensitivity / flagged_sensitivity

Probabilities over 7 sensitivity classes; >0.7 β†’ flagged

has_key_claims

Whether it contains factual claims that need verification (noul probability)

has_engagement_hook

Whether the opening has a strong hook (good for short-form clips)

publish_gate

publish / review / block (see gate logic)

confidence

Calibrated confidence of the sentiment dimension

publish_gate gate logic

  • publish: no sensitivity class > 0.7

  • review: a sensitivity class in 0.7–0.9 β†’ human review before publishing

  • block: any sensitivity class > 0.9 β†’ high risk, block

⚠️ Capability boundary (important): Jev is a decision model β€” it labels / scores, it does not write section headings or summaries. Do chapter segmentation and summarization yourself based on the transcript segments (or plug in a separate generative model). Before any publishing action, check publish_gate + confidence first β€” do not auto-publish just because one probability is high.


Related MCP server: VideoContext MCP

Zero-key / local-first

  • Transcription layer: MEDIA_MCP_MOCK=1 β†’ skip yt-dlp/whisper and return an offline stub (zero-config for CI/demos). Real mode requires yt-dlp and a compiled whisper.cpp on your machine, plus WHISPER_MODEL_PATH configured.

  • Decision layer: no TYPESAFE_API_KEY β†’ automatic Jev mock mode (deterministic offline stub); set it β†’ call the real API.

Environment variables

Variable

Default

Description

TYPESAFE_API_KEY

β€”

Required for real Jev calls; blank falls back to mock

JEV_MCP_MOCK

0

1 forces Jev mock

MEDIA_MCP_MOCK

0

1 skips transcription binaries

YT_DLP_BIN / WHISPER_BIN

yt-dlp / whisper-cli

Binary paths

WHISPER_MODEL_PATH

β€”

Absolute path to the .ggml model file

WHISPER_MODEL

base.en

Model name hint

MEDIA_MCP_TIMEOUT_MS

600000

Transcription timeout

Install engines locally (real transcription)

pip install yt-dlp
git clone https://github.com/ggerganov/whisper.cpp && cd whisper.cpp && make
# download a small model, e.g.:
bash ./models/download-ggml-model.sh base.en
export WHISPER_MODEL_PATH="$PWD/models/ggml-base.en.bin"

Extremely low cost

Transcription runs on your own machine (CPU/GPU, zero call fees); decisions are a single Jev call, tens of thousands of tokens per call. There's no token-usage chart because it's already cheap β€” think of it as "one cheap gate in your content pipeline".


Tests

npm test     # smoke + MCP protocol handshake (dual mock mode, no key needed)

License

MIT

Available Tools

4 tools
analyze_transcriptA

Run a transcript through Jev (System One decision model) for structured analysis: topic membership, sentiment, sensitivity flags (profanity/violence/medical-advice/financial-advice/…), and a publish gate (publish / review / block). The agent gets typed labels, not prose. Always check publish_gate + confidence before publishing.

Inputs: transcript (required), categories.

fetch_mediaA

Download the best audio from a video/audio URL to a local file using yt-dlp. Returns the local path plus title and duration. No API key required.

Inputs: url (required), format.

transcribe_fileC

Transcribe a local audio/video file with whisper.cpp. Returns the transcript text and optional segments.

Inputs: path (required), model, language, translate, +1 more.

transcribe_mediaA

Turn any video/audio URL or local file into a transcript, locally and key-free. URLs are fetched with yt-dlp; local files and fetched audio are transcribed with whisper.cpp. Returns the full transcript text plus optional [start --> end] segments.

Inputs: url_or_path (required), model, language, translate, +1 more.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.7/5.0

    Scored across 4 tools

    Disambiguation3/5

    transcribe_media and transcribe_file overlap for local files, and transcribe_media already handles URL fetching that fetch_media also provides. However, the descriptions make each tool's primary role fairly clear: fetch_media is download-only, transcribe_file is local-only, and transcribe_media is the all-in-one path.

    Naming Consistency5/5

    All tool names follow a consistent verb_noun snake_case pattern: transcribe_media, fetch_media, transcribe_file, analyze_transcript. The naming is predictable and easy to navigate.

    Tool Count4/5

    Four tools is a reasonable size for this audio/video analysis pipeline. The count is only slightly higher than necessary because transcribe_file is mostly redundant with transcribe_media, but it is not bloated.

    Completeness5/5

    The server covers the full workflow: fetch media, transcribe it, and analyze the transcript with the Jev model. There are no obvious missing operations for the stated AV-analysis purpose.

    Related MCP Connectors

    Related MCP Servers