Description
AutoVideo is an instruction Skill for an AI media production workspace. It guides an agent through ideas, scripts, storyboards, multi-shot films, video and image generation or editing, music, voice-overs, sound effects, transcription, media analysis, reusable characters and voices, and task or asset management. The Skill requires the agent to read AutoVideo's live official instructions before using an API or MCP tool. It does not include service credentials or usage credits.
What you get
- Download the source Skill as a ZIP after free acquisition on Finch.
- Includes the original SKILL.md and MIT license.
Usage and examples
- Read the current official AutoVideo agent contract.
- Map the request to documented capabilities and authentication requirements.
- Execute the authorized workflow and track asynchronous tasks.
- Return the resulting media or a documented actionable error.
Use AutoVideo to plan and create a short product video, following the live agent contract and checking usage before paid operations.
The requested media or durable result reference, or a concise actionable error if the documented operation fails.
Requirements and limitations
- An agent runtime that can read the official AutoVideo onboarding page and llms.txt.
- Valid AutoVideo service access for the operations requested; service charges may apply.
- This is an instruction package, not a bundled media-generation engine.
- Capabilities, pricing, and availability depend on the live AutoVideo service.
- No service credentials or generation credits are included.
- Publication checks covered the package structure and source integrity, not a paid end-to-end media run.
- If the official agent page or llms.txt is unavailable, stop before state-changing operations.
- Follow the live contract for task status, authentication, billing, and error handling.
Onchain Identity
Public Metadata
The publisher has not published this metadata publicly.
v1 · Initial Finch encrypted ZIP publication; original Skill instructions preserved.
Files
File verification detailsFor audits and support
Opaque identifiers and fingerprints are hidden from the primary experience but remain available for exact verification.
- Markdown hash
0x97b209120c9ad56a1753bf98759dba10009b85c6b449cdbc7dfe01572910f41c- Package hash
0x1ffe6458991f3337ae463bce6ca8cbd5f84257e1f529423aa7a9b06d6b874c1b