spark-video-episode

65 / 100 · Review before installing

Check before installing. Some of what it can do is expected: a media skill usually calls an AI service over the internet.

By modelstudioai/skills (GitHub) · media · 59 stars · scanned at commit 39b4fcc

One-shot autopilot orchestrator — runs the full spark-video pipeline (screenwriter ↔ director per-scene parallel → render chain-DAG parallel + per-clip review → stitch). User confirms at 4 gates (+ 1 mode gate at start + 1 BGM gate when bgm/ folder detected). Use when the user wants "make me an episode" in one command.

Works with any agent that supports SKILL.md skills.

Partially scanned — some files couldn't be read; we'll retry on the next scan

What this skill can do

Runs commands on your computer

The author does not mention this.

Found in 33 places.

skills/spark-video/.cursor/hooks.json, line 6"command": ".cursor/hooks/spark-video-gate.sh",
skills/spark-video/.cursor/hooks/spark-video-gate.sh, line 56proc = subprocess.run(
skills/spark-video/SKILL.md, line 13SKILL_DIR="$(cd "$(dirname "${BASH_SOURCE[0]:-$0}")" && pwd)"

Show 17 more

skills/spark-video/SKILL.md, line 36export SPARK_VIDEO_PROJECT=<project_id>
skills/spark-video/SKILL.md, line 126./scripts/doctor.sh # bl + ffmpeg + uv present
skills/spark-video/SKILL.md, line 170test -d projects/$SPARK_VIDEO_PROJECT/bgm || \
skills/spark-video/SKILL.md, line 181uv run scripts/scaffold.py cast-init # build cast.json
skills/spark-video/SKILL.md, line 207uv run scripts/storyboard.py compile --mode <drama|narration>
skills/spark-video/SKILL.md, line 215uv run scripts/build_viewer.py # opens viewer.html in browser for review
skills/spark-video/SKILL.md, line 249uv run scripts/build_viewer.py # opens viewer.html — now includes scenes + shots
skills/spark-video/SKILL.md, line 265uv run scripts/render_all.py --reset --ratio 9:16
skills/spark-video/SKILL.md, line 299uv run scripts/build_viewer.py # opens viewer.html — all clips + reviews visible
skills/spark-video/SKILL.md, line 305jq '.[] | {shot: .shot_id, ver: .winner_version,
skills/spark-video/SKILL.md, line 317uv run scripts/stitch.py --crossfade 0.5
skills/spark-video/SKILL.md, line 332uv run scripts/build_viewer.py
skills/spark-video/references/spark-video-cast/SKILL.md, line 24export SPARK_VIDEO_PROJECT=<project_id>
skills/spark-video/references/spark-video-cast/SKILL.md, line 72uv run scripts/scaffold.py cast --name "陆辰"
skills/spark-video/references/spark-video-cast/SKILL.md, line 80./scripts/bl image generate \
skills/spark-video/references/spark-video-cast/SKILL.md, line 105uv run scripts/scaffold.py cast --name "钱夫人" --episode
skills/spark-video/references/spark-video-cast/SKILL.md, line 117uv run scripts/scaffold.py cast-init # merges project + episode tiers

and 13 more places

Downloads and installs other software

The author does not mention this.

Found in 6 places.

skills/spark-video/.qwen/commands/bootstrap.toml, line 1description = "Self-check and auto-install the videogen workspace (ffmpeg / python / pip install -e / .env / Shanyin skills / git hooks)."
skills/spark-video/.qwen/commands/bootstrap.toml, line 12• `ffmpeg -version` — missing on macOS → `brew install ffmpeg`. On Linux,
skills/spark-video/.qwen/commands/bootstrap.toml, line 19- prefer existing `.venv/`: `source .venv/bin/activate && pip install -e .`

Show 3 more

skills/spark-video/.qwen/commands/bootstrap.toml, line 20- else: `python3 -m venv .venv && source .venv/bin/activate && pip install -e .`
skills/spark-video/README.md, line 196- `bl: command not found` → `npm install -g bailian-cli && npx skills add modelstudioai/skills --all -g && bl auth login`
skills/spark-video/README.zh.md, line 187- `bl: command not found` → `npm install -g bailian-cli && npx skills add modelstudioai/skills --all -g && bl auth login`

Changes files outside its own folder

The author does not mention this.

Found in 2 places.

skills/spark-video/README.md, line 215rm -rf ~/.claude/skills/spark-video
skills/spark-video/README.zh.md, line 206rm -rf ~/.claude/skills/spark-video

Sends data to these websites: dashscope.aliyuncs.com

The author does not mention this. Expected for a media skill.

Found in 23 places.

skills/spark-video/api-references/dashscope/image/wan-text2imageV2.md, line 150| #### 请求参数 | ## 查询任务结果将`{task_id}`完整替换为上一步接口返回的`task_id`的值。`task_id`查询有效期为24小时。 ``` curl -X GET https://dashscope.aliyuncs.com/api/v1/tasks/{task_id} \\ --hea
skills/spark-video/api-references/dashscope/image/wan-text2imageV2.md, line 192| #### 请求参数 | ## 文生图 ``` curl -X POST https://dashscope.aliyuncs.com/api/v1/services/aigc/text2image/image-synthesis \\ -H 'X-DashScope-Async: enable' \\ -H "Au
skills/spark-video/api-references/dashscope/image/wan-text2imageV2.md, line 233| #### 请求参数 | ## 查询任务结果请将`86ecf553-d340-4e21-xxxxxxxxx`替换为真实的task\\_id。 > 若使用新加坡地域的模型，需将base\\_url替换为https://dashscope-intl.aliyuncs.com/api/v1/tasks/86ecf553-

Show 17 more

skills/spark-video/api-references/dashscope/image/wan-text2imageV2.md, line 413status = ImageGeneration.fetch(task=task, api_key=api_key)
skills/spark-video/api-references/dashscope/image/wan-text2imageV2.md, line 566f.write(requests.get(result.url).content)
skills/spark-video/api-references/dashscope/image/wan-text2imageV2.md, line 663f.write(requests.get(result.url).content)
skills/spark-video/api-references/dashscope/image/wan-text2imageV2.md, line 671status = ImageSynthesis.fetch(task=task, api_key=api_key)
skills/spark-video/api-references/dashscope/image/wan-text2imageV2.md, line 1104ImageSynthesisResult result = is.fetch(taskId, apiKey);
skills/spark-video/api-references/dashscope/tts/cosyvoice-tts.md, line 17| ## **请求体** | ## 非流式 ``` curl -X POST https://dashscope.aliyuncs.com/api/v1/services/audio/tts/SpeechSynthesizer \\ -H "Authorization: Bearer $DASHSCOPE_API_KE
skills/spark-video/api-references/dashscope/video/happyhorse/happyhorse-1.0-i2v.md, line 89| #### 请求参数 | ## 查询任务结果将`{task_id}`完整替换为上一步接口返回的`task_id`的值。`task_id`查询有效期为24小时。 ``` curl -X GET https://dashscope.aliyuncs.com/api/v1/tasks/{task_id} \\ --hea
skills/spark-video/api-references/dashscope/video/happyhorse/happyhorse-1.0-r2v.md, line 88| #### 请求参数 | ## 查询任务结果将`{task_id}`完整替换为上一步接口返回的`task_id`的值。`task_id`查询有效期为24小时。 ``` curl -X GET https://dashscope.aliyuncs.com/api/v1/tasks/{task_id} \\ --hea
skills/spark-video/api-references/dashscope/video/happyhorse/happyhorse-1.0-t2v.md, line 89| #### 请求参数 | ## 查询任务结果将`{task_id}`完整替换为上一步接口返回的`task_id`的值。`task_id`查询有效期为24小时。 ``` curl -X GET https://dashscope.aliyuncs.com/api/v1/tasks/{task_id} \\ --hea
skills/spark-video/api-references/dashscope/video/happyhorse/happyhorse-1.0-video-edit.md, line 88| #### 请求参数 | ## 查询任务结果将`{task_id}`完整替换为上一步接口返回的`task_id`的值。`task_id`查询有效期为24小时。 ``` curl -X GET https://dashscope.aliyuncs.com/api/v1/tasks/{task_id} \\ --hea
skills/spark-video/api-references/dashscope/video/wan/wan-image2video.md, line 95| #### 请求参数 | ## 查询任务结果将`{task_id}`完整替换为上一步接口返回的`task_id`的值。`task_id`查询有效期为24小时。 ``` curl -X GET https://dashscope.aliyuncs.com/api/v1/tasks/{task_id} \\ --hea
skills/spark-video/api-references/dashscope/video/wan/wan-image2video.md, line 264status = VideoSynthesis.fetch(task=rsp, api_key=api_key)
skills/spark-video/api-references/dashscope/video/wan/wan-reference2video.md, line 88| #### 请求参数 | ## 查询任务结果将`{task_id}`完整替换为上一步接口返回的`task_id`的值。`task_id`查询有效期为24小时。 ``` curl -X GET https://dashscope.aliyuncs.com/api/v1/tasks/{task_id} \\ --hea
skills/spark-video/api-references/dashscope/video/wan/wan-reference2video.md, line 288status = VideoSynthesis.fetch(task=rsp, api_key=api_key)
skills/spark-video/api-references/dashscope/video/wan/wan-text2video.md, line 90| #### 请求参数 | ## 查询任务结果将`{task_id}`完整替换为上一步接口返回的`task_id`的值。`task_id`查询有效期为24小时。 ``` curl -X GET https://dashscope.aliyuncs.com/api/v1/tasks/{task_id} \\ --hea
skills/spark-video/api-references/dashscope/video/wan/wan-text2video.md, line 243status = VideoSynthesis.fetch(task=rsp, api_key=api_key)
skills/spark-video/api-references/dashscope/video/wan/wan-text2video.md, line 500VideoSynthesisResult result = is.fetch(taskId, apiKey);

and 3 more places

Safer alternatives

Low-risk skills in the same category (media) that look related. Check that one does what you need.

Other copies in this repository

modelstudioai/skills has 1 more folder with a skill named spark-video-episode. This page shows the lowest-scoring copy.

skills/spark-video/references/spark-video-episode 100/100 · Low risk

Install

Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.

if [ -e ~/.agents/skills/'spark-video-episode' ] || [ -L ~/.agents/skills/'spark-video-episode' ]; then echo 'A skill folder named spark-video-episode already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/modelstudioai/skills' "$d" && git -C "$d" checkout --quiet '39b4fcc94df3072baa06d6431039e3a91716e61b' -- 'skills/spark-video' && mkdir -p ~/.agents/skills && cp -R "$d"/'skills/spark-video' ~/.agents/skills/'spark-video-episode' && rm -rf "$d"; fi

Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'spark-video-episode' ~/.claude/skills/'spark-video-episode' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.

Badge for your README

The badge for modelstudioai/skills shows the score of its lowest-scoring skill.

SkillTrust badge for modelstudioai/skills

[![SkillTrust](https://zalavijaysinh.github.io/skilltrust/badges/modelstudioai/skills.svg)](https://zalavijaysinh.github.io/skilltrust/repos/modelstudioai/skills/)

Score history

Date	Version	Score
2026-11-13	`39b4fcc`	65
2026-11-07	`18d749e`	65