FireRed-OpenStoryline is an AI video editing agent that uses natural language, LLM planning, and tool orchestration to assist in story-driven editing. It includes agent skills, OpenClaw/Claude Code integration, and a local MCP server with a docker option.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
FireRed-OpenStoryline is an AI video editing agent that transforms manual editing into intention-driven directing through natural language interaction, LLM-powered planning, and precise tool orchestration. It supports reusable editing workflows via Skills, and includes OpenClaw/Claude Code integration paths.
How it works
The project exposes an Agent framework with Skills such as openstoryline-install and openstoryline-use to install and run the editing workflow. It provides a Model Context Protocol (MCP) and video processing Nodes, plus a skills library for agents. OpenClaw and Claude Code integrations are described with example commands to install and activate skills.
Getting started
Install and setup steps include:
- Clone the repository and create a Python/conda environment.
- Use either automatic or manual resource installation scripts.
- Start MCP server and the conversation interface via CLI or web interface.
Commands shown in the README include, for example:
git clone https://github.com/FireRedTeam/FireRed-OpenStoryline.git
cd FireRed-OpenStoryline
conda create -n storyline python=3.11
conda activate storyline
sh build_env.sh
PYTHONPATH=src python -m open_storyline.mcp.server
python cli.py
uvicorn agent_fastapi:app --host 127.0.0.1 --port 8005
Recent releases
No releases listed in the latest section.
Traction
Statistics present: 3183 stars.
Behind the repo
The repository is from FireRedTeam and features integration points with Hugging Face and ModelScope, and a multi-component architecture (src/open_storyline, docs, resources, prompts, etc.).
Caveats
License: Apache-2.0. Language: Python. Created 2026-02-07; last_push 2026-07-31. Several resource download/install steps depend on external scripts (download.sh, build_env.sh) and platform-specific commands for macOS/Linux/Windows. Docker image versions shown: v1.0.1.