← Back to Skills Marketplace
Whisk
by
vcarolxhberger
· GitHub ↗
· v1.0.1
· MIT-0
114
Downloads
0
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install whisk
Description
Drop a video and describe the look you're after — Whisk reads your footage and remixes its visual style, pacing, and mood on the fly. Whether you want to tur...
Usage Guidance
This skill appears to do what it says: it calls the NemoVideo API to analyze and remix uploaded video. Before installing: 1) Be aware the skill requires an API token (NEMO_TOKEN). Treat that token like a password and only provide a token scoped/limited for this use; rotate it if needed. 2) The skill constructs a claim link that embeds the token as a URL parameter — avoid sharing that link and prefer provider flows that don't expose tokens in URLs. 3) The skill will write a client_id file to ~/.config/nemovideo/ if absent; this is benign but review that file if you care about local artifacts. 4) Only upload video files you are authorized to process. If you need stronger assurances, ask the vendor for token-scoping options, shorter token lifetimes, or an alternative claim flow that doesn't put tokens in URLs.
Capability Analysis
Type: OpenClaw Skill
Name: whisk
Version: 1.0.1
The 'whisk' skill is a legitimate integration for the NemoVideo AI video editing service. It provides instructions for an AI agent to manage video remixing tasks, including session creation, file uploads, and status tracking via the 'mega-api-prod.nemovideo.ai' backend. The skill's use of local configuration in '~/.config/nemovideo/' and its authentication flow (using environment variables or fetching anonymous tokens) are standard for this type of service and align with its stated purpose.
Capability Assessment
Purpose & Capability
Name/description (video style remixer) match the declared dependency (a single NEMO_TOKEN) and the SKILL.md which describes calls to the NemoVideo API. The required config path (~/.config/nemovideo/) and client_id UUID are plausibly needed for anonymous token acquisition and session creation.
Instruction Scope
Instructions focus on creating a session, uploading video assets, and streaming SSE messages to the NemoVideo API. A noteworthy point: the SKILL.md directs building a claim link that embeds the token as a URL query parameter (https://nemovideo.com/workspace/claim?token=$TOKEN&...), which can leak the token via logs, browser history, or if the link is shared. Otherwise the instructions do not ask the agent to read unrelated system files or other credentials.
Install Mechanism
This is an instruction-only skill with no install spec and no code files — nothing is written to disk by an installer. The only file-write action the instructions mention is creating a client_id under ~/.config/nemovideo/ if needed, which is reasonable for anonymous auth.
Credentials
Only a single credential (NEMO_TOKEN) is required and declared as primaryEnv, which is proportionate. Caveat: the created claim URL includes the token in a query parameter, which increases risk of token exposure. The skill also relies on a local client_id config file when obtaining anonymous tokens — this is consistent with its described behavior.
Persistence & Privilege
The skill is not force-included (always: false) and can be invoked by the user. It does not request elevated platform privileges or modify other skills' configuration. It does instruct creating a small client config under the user's homedir, which is normal for a client identifier.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install whisk - After installation, invoke the skill by name or use
/whisk - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
- Whisk's focus shifted from recipe/ingredient detection to AI-powered video style remixing.
- Updated description, display name, and examples to reflect new creative editing features (style transfer, tone/pacing changes, rhythm-matched cuts).
- Changed target audience from home cooks/food creators to content creators, marketers, and social editors.
- Metadata "emoji" changed from 🥄 to 🌀 in alignment with new creative video purpose.
- Instructions, workflows, and introduction revised for video transformation instead of culinary analysis.
v1.0.0
Whisk 1.0.0 — AI-Powered Video Ingredient & Recipe Identification Tool
- Initial release: Turn any cooking video into an interactive, step-by-step recipe workflow.
- Automatically identifies ingredients, quantities, cooking methods, and recipe steps from uploaded cooking videos (mp4, mov, avi, webm, mkv).
- Simple setup: Handles free, anonymous API auth and session management in the background.
- Supports direct questions—ingredient scan, recipe extraction, pantry mapping—with seamless video analysis.
- Clear API integration with session management, credits check, export, and rich error handling.
- Built for home cooks, food creators, and culinary enthusiasts seeking instant, actionable kitchen guidance from their videos.
Metadata
Frequently Asked Questions
What is Whisk?
Drop a video and describe the look you're after — Whisk reads your footage and remixes its visual style, pacing, and mood on the fly. Whether you want to tur... It is an AI Agent Skill for Claude Code / OpenClaw, with 114 downloads so far.
How do I install Whisk?
Run "/install whisk" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Whisk free?
Yes, Whisk is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Whisk support?
Whisk is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Whisk?
It is built and maintained by vcarolxhberger (@vcarolxhberger); the current version is v1.0.1.
More Skills