Preface

The core concept of DeepSeek Harness (DSH) is “everything is a plugin.” For technical professionals or creators who need to build content workflows, the ability to directly call external tools in DSH sessions is key to improving efficiency. FrameFactory is a content production tool, and this plugin provides a bridge to connect its public capabilities to DSH.

Plugin Positioning

This is a minimal viable DeepSeek Harness plugin maintained by anywhere-labs. It wraps FrameFactory’s public capabilities into a callable skill, allowing developers to directly generate content briefs in the DSH environment.

Core Capabilities

The plugin currently provides the following features:

  • Topic breakdown
  • Social media title candidates
  • Short video script structure
  • Visual planning and asset sourcing suggestions
  • Verification items and pre-publish QC checklist
  • A dependency-free JSON brief example tool

Installation and Enablement

After installing the latest DSH, run the following command in the target profile:

dsh plugin --profile web add github:anywhere-labs/framefactory-plugin

After running the installation command, enter the profile:

dsh web

Usage Example

After successful installation, start a new session and enter a prompt similar to the following:

使用 FrameFactory，帮我把"AI Agent 如何改变自媒体工作流"做成一条短视频内容 brief。

The model will invoke the framefactory-content skill and output the title, script, visual plan, and QC checklist.

For local development verification, you can also use the provided script directly:

node tools/render-brief.mjs "AI Agent 如何改变自媒体工作流"

Notes

This is a public MVP plugin for FrameFactory. It does not include the team’s non-public production-level writing process cards, commercial asset libraries, model keys, or internal QC implementations. The plugin runs with DSH process permissions, so review the source code and license before installation.

Conclusion

This plugin adds a standardized interface for content generation to the DSH ecosystem, making it suitable for developers who want to integrate FrameFactory capabilities into automated workflows. The related code and documentation are available in the GitHub repository.