Video Skill
v0.1.2Run the video-skill pipeline to convert narrated videos into structured step data and enriched timeline-ready outputs. Use when a user asks to process a vide...
⭐ 1· 327·2 current·2 all-time
byMichael Gold@michaelgold
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
medium confidencePurpose & Capability
Name/description (convert narrated videos into steps and enriched outputs) aligns with the code and CLI commands. The required binaries (uv, ffmpeg, python3) are reasonable for a Python CLI that uses ffmpeg and the uv packaging/runtime. The included scripts, docker-compose, and model-bootstrapping are appropriate for a self-hosted model-backed pipeline.
Instruction Scope
SKILL.md and the CLI instruct the agent/operator to run transcription, chunking, extraction, frame sampling, enrichment, and markdown rendering. The instructions direct the tool to call configured model provider endpoints and to read and base64-encode image files (frames) and include them in model requests — expected for VLM-based enrichment but important to note: large binary image payloads will be sent to whatever provider URL is configured. SKILL.md contains a minor contradiction (claims 'no repo clone required' while showing commands that assume a local repo).
Install Mechanism
The registry lists no automated install spec (instruction-only). The bundle nevertheless contains many source files and helper scripts. This is not itself dangerous, but the provided scripts (scripts/bootstrap_models.sh) will download large model binaries from Hugging Face via the HF CLI and docker-compose references GHCR images — both reasonable for a local/self-hosted setup but require trust in those sources and will write substantial data to disk.
Credentials
The skill does not require any environment variables by default (config.example.json uses api_key_env:null). The code supports provider API keys if configured, which is appropriate for calling model endpoints. There are no unrelated credentials requested in the manifest. Note: if you set provider.api_key_env to point to an env var, that env var will be used to authenticate requests to the configured model endpoints — so only set keys for providers you trust.
Persistence & Privilege
always:false and no install spec means the skill does not request forced persistent inclusion or elevated platform privileges. It performs normal file I/O (reads/writes JSONL, writes frames/clips, runs ffmpeg subprocesses) and spawns subprocesses; this is expected behavior for a CLI tool of this scope.
Assessment
This package appears coherent for its stated purpose, but review these before running: 1) Provider trust: the enrichment steps will send transcript text and base64-encoded frame images to whichever base_url you configure for 'reasoning' and 'vlm' — only point those at services you control or trust. 2) Model downloads & docker: bootstrap_models.sh will download large model files (requires HF CLI and an authenticated account) and the docker-compose file pulls images from GHCR — verify sources and run in a machine with sufficient disk/GPU or use an isolated VM/container. 3) Local commands & subprocesses: the tool invokes ffmpeg and subprocess.run (clip extraction); run on files you trust and consider limiting permissions/using a sandbox. 4) Config review: config.example.json leaves api_key_env null; if you populate api_key_env make sure env vars are set appropriately and contain only credentials for intended providers. 5) Minor docs inconsistency: SKILL.md says 'no repo clone required' but many commands expect a local repo — follow the README/INSTRUCTIONS for correct setup. If you need higher assurance, review the remaining truncated source files (settings and any network code) and run the pipeline in a disposable container before using with sensitive data.Like a lobster shell, security has layers — review code before you run it.
latestvk97egkcj7zr5hasc72bcnp27j9823hyq
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
Binsuv, ffmpeg, python3