Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/JosephOIbrahim/Comfy-CozyWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/josephoibrahim/comfy-cozy/analyze-video)<a href="https://agentmods.dev/commands/josephoibrahim/comfy-cozy/analyze-video"><img src="https://agentmods.dev/badge/commands/josephoibrahim/comfy-cozy/analyze-video/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/josephoibrahim/comfy-cozy/analyze-video"><img src="https://agentmods.dev/badge/commands/josephoibrahim/comfy-cozy/analyze-video.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.00984 |
| Opus 5 | $0.00000 | $0.00492 |
| Sonnet 5 | $0.00000 | $0.00197 |
| Haiku 4.5 | $0.00000 | $0.00098 |
Grade A, and why
analyze-video scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -L -o workspace/reference/input_video.mp4 "$VIDEO_URL" How it starts
The opening of the file, as written. The whole thing — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/project:analyze-video — ANALYST Expert
You are the ANALYST expert in the Video Recreation Agent Team.
Your job: Take a reference video and produce a structured storyboard that the rest of the team can build from.
Input
$ARGUMENTS — A video URL or local file path
Steps
1. Download/Locate the video
# If URL, download it
curl -L -o workspace/reference/input_video.mp4 "$VIDEO_URL"
# If local path, copy to workspace
cp "$VIDEO_PATH" workspace/reference/input_video.mp4
2. Get video metadata
ffprobe -v quiet -print_format json -show_format -show_streams workspace/reference/input_video.mp4
Extract: duration, fps, resolution, codec, audio presence.
3. Detect scene boundaries
# Extract frames at scene changes (threshold 0.3 = moderate sensitivity)
mkdir -p workspace/keyframes
ffmpeg -i workspace/reference/input_video.mp4 \
-vf "select='gt(scene,0.3)',showinfo" \
-vsync vfr \
workspace/keyframes/scene_%04d.png 2>&1 | grep showinfo
Parse the showinfo output to get timestamps of each scene change.
If fewer than 3 scenes detected, lower threshold to 0.2. If more than 12, raise to 0.4.
4. Extract keyframe pairs (first + last frame per segment)
For each detected segment:
# First frame of segment
ffmpeg -ss {start_time} -i workspace/reference/input_video.mp4 -frames:v 1 workspace/keyframes/seg_{N}_first.png
# Last frame of segment
ffmpeg -ss {end_time - 0.04} -i workspace/reference/input_video.mp4 -frames:v 1 workspace/keyframes/seg_{N}_last.png
5. Analyze each keyframe with Vision
For each keyframe, describe:
- Shot type: Wide, medium, close-up, overhead, POV
- Camera: Fisheye, standard, tracking, static, handheld
- Style: Color grade, overlays, effects (scanlines, timestamps, bounding boxes)
- Subject: What/who is in frame, their pose, position
- Mood: Emotional tone of the shot
- Lighting: Key light direction, color temperature, contrast
6. Produce storyboard JSON
Save to workspace/storyboard.json with this exact schema:
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago First seen · 114 lines · 0 tokens per session scan A 2e83b8432c75
analyze-video is a command published in the GitHub repository JosephOIbrahim/Comfy-Cozy (24 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 984 tokens. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
director
Direct a short film from a story. Generates scenes, frames, and video clips.
compare
Diff two ComfyUI workflows to see what changed.
install
Install a ComfyUI custom node pack.
gallery
Browse and inspect generated ComfyUI outputs.
viz
Visualize a ComfyUI workflow as a mermaid diagram.
convert
Convert between ComfyUI UI format and API format workflows.