Uloop Record Video logo

Uloop Record Video

Community
hatayama
uloop-record-video

Record the Game View or any Editor window (Scene, Inspector, ...) to an .mp4 or .webm video. Use when a screenshot is not enough: motion, animation, physics, or a gameplay sequence.

Overview

Publisherhatayama
Repositoryunity-cli-loop
Skill nameuloop-record-video
Stars
578
Forks
49
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by hatayama on GitHub. Read the source before you install it.

Installation

Install the Uloop Record Video AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/hatayama/unity-cli-loop/tree/main/.agents/skills/uloop-record-video
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/hatayama/unity-cli-loop.git /tmp/unity-cli-loop
mkdir -p .claude/skills
cp -r /tmp/unity-cli-loop/.agents/skills/uloop-record-video .claude/skills/uloop-record-video
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Uloop Record Video in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Uloop Record Video on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Uloop Record Video is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Task

Record the Unity Game View while Play Mode is running, or any Editor window with --window-name, then hand the file path to the user or inspect frames from it.

Workflow

  1. For a Game View recording, ensure Play Mode is running (uloop control-play-mode --action Play) and the Game View is open; start fails in Edit Mode. With --window-name the recording targets that Editor window instead and needs no Play Mode.
  2. uloop record-video --action start [options]. It returns immediately; encoding continues inside the Editor.
  3. Drive the scene with other uloop commands (simulate-keyboard, simulate-mouse-input, replay-input, ...). Do not run uloop compile mid-recording: a compile that runs auto-stops and finalizes the file (when Unity's Script Changes While Playing is Recompile After Finished Playing, the compile is refused instead and the recording continues). Exiting Play Mode also auto-stops a Game View recording; a --window-name recording keeps running when Play Mode stops.
  4. uloop record-video --action stop. The file is playable only after this call (or after an auto-stop).
  5. Read OutputPath from the JSON and use exactly that path. The output directory holds earlier recordings too (the newest 20 per extension are kept), so ls -t can pick a stale file.
  6. To inspect the content yourself, extract stills with ffmpeg (e.g. ffmpeg -i "<OutputPath>" -vf fps=1 frames_%03d.png) and view the PNGs. Otherwise report the path and duration to the user.

Tool Reference

bash
uloop record-video --action start [--frame-rate <fps>] [--max-duration-seconds <sec>] [--resolution-scale <0.1-1.0>] [--quality <low|medium|high>] [--window-name <name>] [--match-mode <exact|prefix|contains>] [--output-path <file>]
uloop record-video --action status
uloop record-video --action stop

Parameters

ParameterTypeDefaultDescription
--actionenumstartstart - begin recording, stop - finalize the file, status - report progress
--frame-rateinteger30Output video fps. Valid range 1–60. Frames are paced by wall-clock time; when the Editor renders slower than this, the previous frame is repeated. Used by start only.
--max-duration-secondsinteger60Auto-stop safety limit in seconds. Valid range 1–600. Used by start only.
--resolution-scalenumber1.0Resolution scale (0.1 to 1.0) applied to the Game View size before encoding. 0.5 cuts file size and encoding cost to about a quarter. Used by start only.
--qualityenummediumEncoder bitrate preset: low, medium, or high. Used by start only.
--window-namestringemptyEditor window title to record instead of the Game View (for example Scene, Inspector, Console). Empty records the Game View and requires Play Mode. A window recording does not require Play Mode, brings the tab to the front, repaints it every frame, and keeps running when Play Mode stops. Used by start only.
--match-modeenumexactWindow title matching for --window-name: exact, prefix, or contains (case-insensitive). Used by start only.
--output-pathstringemptyOutput file path. Empty uses .uloop/outputs/Videos/gameview_<yyyyMMdd_HHmmss_fff>.mp4 (.webm on Linux). Extension must be .mp4 (H.264) or .webm (VP8). Linux rejects .mp4. Used by start only.

Actions

ActionBehaviorTypical use
startValidates, opens the encoder, returns. One recording at a time; a second start fails until stop.Begin capture before driving input
statusReports live counters. When idle, reports the most recent auto-stopped recording.Check progress or find out why a recording ended
stopFinalizes the file and returns the final counters. When nothing is recording, returns the last auto-stopped recording once, then "No recording is in progress."End capture

Output

Returns JSON containing:

  • Success: Whether the request completed.
  • Message: Human-readable status.
  • Action: Echoes the executed action.
  • IsRecording: Whether a recording is active after this call.
  • OutputPath: Absolute path of the video file. Open this path; do not search the directory. A window recording uses the window_ name prefix instead of gameview_.
  • Width / Height: Encoded resolution after --resolution-scale and even rounding (0 when none).
  • FrameRate: Output fps (0 when none).
  • Quality: Bitrate preset in use.
  • EncodedFrameCount: Frames written to the encoder.
  • SkippedFrameCount: Frame slots that could not be captured (Game View closed or resized, or encoder refused a frame). The video keeps its timeline; skipped slots are simply missing.
  • ElapsedSeconds: Seconds since start (frozen after stop).
  • StoppedBy: Why the recording ended — "cli", "max-duration", "play-mode-exit", "window-closed", "assembly-reload", or "editor-quit". Omitted while recording.

Interpreting results

  • Success: false with Message "A recording is already in progress" → run stop first, then start again.
  • Success: false with "Play Mode view RenderTexture is not available" → open the Game View tab (uloop focus-window) and make sure a camera renders.
  • SkippedFrameCount growing while EncodedFrameCount stays flat → the Game View is closed, hidden, or resized. Restore it; recording resumes without restarting.
  • EncodedFrameCount far below ElapsedSeconds × FrameRate with few skips → the Editor is unfocused and throttling draws; run uloop focus-window before the next recording.
  • Success: false with "Window '...' not found" → pick the right title from the Open windows: list in the message, or use --match-mode prefix.
  • StoppedBy is not "cli" → the recording ended on its own; the file is still valid up to that point.

Notes

  • Odd Game View sizes are rounded down to even for H.264 (for example 1286×723 → 1286×722; the last pixel row/column is dropped).
  • The recording is wall-clock paced, so Play Mode pause records a still image, not a gap.
  • A window recording survives Play Mode stopping, but entering Play Mode with domain reload enabled (Unity's default) ends it with StoppedBy: "assembly-reload".
  • Default output is H.264 .mp4 on macOS/Windows and VP8 .webm on Linux. An explicit .webm path uses VP8 on every host.
  • Default recordings under .uloop/outputs/Videos/ keep only the newest 20 files per extension; a custom --output-path is never pruned.

Frequently asked questions

What does the Uloop Record Video AI skill do?

Record the Game View or any Editor window (Scene, Inspector, ...) to an .mp4 or .webm video. Use when a screenshot is not enough: motion, animation, physics, or a gameplay sequence.

Why use Uloop Record Video on TypingMind?

Because you install it once and use it with any model. Uloop Record Video is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Uloop Record Video in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/hatayama/unity-cli-loop/tree/main/.agents/skills/uloop-record-video. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Uloop Record Video?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Uloop Record Video?

As many as you like. As long as a model supports skills, you can use Uloop Record Video with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Uloop Record Video AI skill free?

Yes. It is published on GitHub by hatayama under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇