dsh-video-understand
DeepSeek Harness video-understanding tool for Bilibili links, BV IDs, and local videos.
Install
No clear install command was found in the current repository evidence.
Plugin Overview & Capabilities
AI-assisted organization based on the public repository snapshot. The content must be grounded in source evidence and does not replace compatibility or security verification.
Key Capabilities
- Registers the `video_understand` tool for DSH agents.
- Accepts Bilibili URLs, BV IDs, and absolute paths to local video files.
- Produces JSON including video metadata, duration, token-compression percentage, cost, and answers.
- Uses a local L0 pipeline for ASR, scene classification, and motion detection.
- Offers L1 and L2 visual-analysis modes that send JPEG video frames to the MiMo API.
- Creates an isolated local Python environment and installs core dependencies on first use.
- Provides a `doctor` CLI for environment checks and automated repair.
Useful For
- Summarizing a Bilibili or local video for an agent workflow.
- Asking targeted questions about video content.
- Extracting a compact information layer before LLM-based analysis.
- Using frame-based visual analysis for clothing, posture, color, or time-window evidence.
Who It Fits
- DeepSeek Harness users building video-analysis agent workflows.
- Users analyzing Bilibili or local video files.
- Developers who need a DSH tool with local-first video processing.
Documented Limitations
- Requires ffmpeg and an LLM API key.
- L1 and L2 send JPEG video frames to the MiMo API for VLM analysis.
- The optional semantic layer using torch, CLIP, and YOLO requires approximately 2 GB of dependencies.
- Bilibili downloading may require configuration of `BILI_DOWNLOAD_SCRIPT`.
- L0 does not provide the visual-detail analysis available in L1 or L2.
DSH Compatibility
Version-specific runtime evidence collected by DSH Plugin. A missing result means we have not tested that combination yet.
Security Signals
Objective signals discovered from package metadata and source inspection. These are not a guarantee that a plugin is safe.
package.json declares 0 runtime, 0 development, 0 peer, and 0 optional dependencies.
package.json declares DSH bundle metadata.
GitHub did not report an SPDX license for this repository snapshot.
A root package.json was captured and can be inspected by the registry.
Public GitHub source metadata is available for this registry snapshot.
Source & Registry Notes
Public provenance, Registry classification, and the latest source check for this entry, kept separate from runtime verification.
- Source repository
- ilps2/dsh-video-understand
- Registry source
- GitHub · dsh-plugin topic
- Registry classification
- Plugin
- Source checked
- 29a4698 · 2026-09-14
This project is independently indexed from public source information. DSH Plugin is not affiliated with DeepSeek or the plugin author. Always check the author repository before installation.
Repository Activity
- GitHub stars
- 8GitHub stars
- Forks
- 0
- Open issues
- 0
- Last commit
- 2026-08-24
- Last release
- No release detected
Related DSH Plugins
Ranked by overlapping capabilities, use cases, plugin type, categories, and DSH profile.
Vision routing and pixel-level visual tool suite for DeepSeek Harness agents with built-in free fallbacks.
View plugin →DeepSeek Harness web-search plugin that adds web search, X search, and focused page fetching with structured evidence.
View plugin →Browser automation plugin enabling DeepSeek Harness agents to interact with logged-in browser sessions safely via dedicated Agent Windows.
View plugin →Image generation and iterative editing plugin for DeepSeek Harness supporting cloud providers and local ComfyUI.
View plugin →