flux-3-video — Claude Code skill: what it does, how to install
Routes general FLUX 3 video requests to the exact sub-skills handling prompt repair, visual staging, audio, and generation.
MIT 185 installs ★ 138 updated 2026-09-09
npx skills add black-forest-labs/skills --skill flux-3-video -a claude-codeWhat flux-3-video does
This skill acts as a router for the FLUX 3 video pipeline. Instead of running everything at once, it breaks down a request and delegates work to specific sub-skills. It assigns brief diagnosis to flux-3-prompt-doctor, camera staging to flux-3-cinematic-inserts, reference-based workflows to flux-3-keyframes-continuation, and rendering to flux-3-generate.
Each handoff carries the selected mode (t2v, i2v, or v2v), delivery constraints, camera directives, and identified risks. Full rendering takes minutes. Because of that, the workflow suggests testing drafts before running full-length jobs. Deterministic requirements like exact typography or frame-accurate audio sync require post-production editing rather than model generation.
What it helps with
- route video generation prompts to the correct FLUX 3 sub-skills
- pass shot structure, camera settings, and aspect ratio across pipeline stages
- select between pure text-to-video and image-to-video continuation
- flag constraints that require external post-production instead of generation
Example prompts for the agent
- Route this video brief to create a text-to-video clip with background audio in FLUX 3
- Set up a pipeline to extend this starting frame into a 5-second video
- Plan the sub-skills needed to produce a commercial video using FLUX 3
How to install flux-3-video
1Skills CLI (Vercel)
npx skills add black-forest-labs/skills --skill flux-3-video -a claude-code2Manually: put the files in place
~/.claude/skills/flux-3-video/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/flux-3-video/
1Skills CLI (Vercel)
npx skills add black-forest-labs/skills --skill flux-3-video -a codex2Manually: put the files in place
.agents/skills/flux-3-video/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add black-forest-labs/skills --skill flux-3-video -a cursor2Manually: put the files in place
.cursor/skills/flux-3-video/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install black-forest-labs/skills skills/flux-3-video2Manually: put the files in place
.github/skills/flux-3-video/SKILL.md1Skills CLI (Vercel)
npx skills add black-forest-labs/skills --skill flux-3-video -a gemini-cli2Platform CLI
gemini skills install https://github.com/black-forest-labs/skills --path skills/flux-3-videoManually: put the files in place
.devin/skills/flux-3-video/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add black-forest-labs/skills --skill flux-3-video -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about flux-3-video
How to install flux-3-video in Claude Code?
Run npx skills add black-forest-labs/skills --skill flux-3-video -a claude-code. The file will be stored in ~/.claude/skills/flux-3-video/SKILL.md.
What does flux-3-video do?
It directs FLUX 3 requests to the minimal set of specialized skills needed for diagnostics, direction, audio, or rendering.
Can flux-3-video handle frame-accurate audio syncing or exact typography?
No. The router treats deterministic elements like exact typography and frame-accurate sync as tasks for post-production, not raw generation.
Source text
SKILL.md · MIT
---
name: flux-3-video
description: Use when routing a general FLUX 3 video request. Select only the specialists needed for diagnosis, direction, audio, or execution.
metadata:
author: Black Forest Labs
version: "1.0.0"
tags: flux, flux-3, bfl, video-generation, routing
---
# FLUX 3 Video Router
Route a request to the smallest set of specialists that can complete it. When the request
already names one narrow job, use that specialist directly.
| User intent | Skill |
| --- | --- |
| Diagnose or repair a brief before generation | `flux-3-prompt-doctor` |
| Write a text-to-video prompt for a new shot | `flux-3-cinematic-inserts` |
| Build from supplied images or video | `flux-3-keyframes-continuation` |
| Direct dialogue, voiceover, ambience, effects, or music | `flux-3-audio-dialogue` |
| Give a shot a period or archival look | `flux-3-archival-formats` |
| Submit, poll, enhance, download, or technically review | `flux-3-generate` |
| Build a finished product ad with voiceover | `flux-3-product-ads` |
Default sequence, skipping completed stages:
```text
Prompt Doctor (until READY; owns the shot-structure decision)
→ Cinematic Inserts OR Keyframes & Continuation (one, chosen by whether
supplied media must constrain the result)
→ Audio & Dialogue (only when sound matters)
→ Generate (only when execution is requested)
```
Draft before committing anything unproven: a full render takes minutes, and the chosen
draft replays at full quality.
A handoff between skills carries: outcome, the route (`mode` plus its media field:
`t2v`, `i2v`/`keyframes`, `v2v`/`start_video`, or `draft_enhance`/`draft_cache`), the
one required action, source roles, invariants, delivery (duration, aspect, resolution,
shot structure), camera, audio, text strategy, and risks. It is complete when the
receiving skill can act without inventing a consequential choice.
Deterministic requirements (exact typography, frame-accurate sync, final mixing) need an
explicit post-production plan, not a generation attempt.
Reference: [docs.bfl.ai](https://docs.bfl.ai)
© Black Forest Labs · MIT · black-forest-labs/skills