Skip to content
Recaply.
Sign inSubmit a tool

DirectorySkills

Full size
Run 1Objects detected in a photo“Detect "Plant 91%" and "Pot 96%" in our plant photo.”
Runs in
  • Claude Code
  • Codex
  • Cursor
  • Gemini CLI
  • GitHub Copilot
  • OpenCode
Generates with
Your agent's model
Tested with
Claude Code 2.1.322Claude Sonnet 5.51 real run, 10 Nov
Needs
Node 22+Chromeffmpeg
Licence
Recaply licence

Objects Detected in a Photo

Detection boxes draw around things in a photo with labels and scores.

$9.99once, yours to keep

Every premium skill and prompt, one plan$19.99 a month. Cancel any time. What you get when you buy

Payments are non-refundable. See our Terms of service.

Boxes draw around objects in a photo one by one, each with a label and a confidence score. A 4 to 10 second clip in 16:9, 9:16, 1:1 or 4:5, as an MP4 on its own or as one scene of a longer launch film. Renders locally with Chrome and ffmpeg: no video model, no key and no cost. Use when a vision, scanning, inventory or AI camera product shows what it recognises.

By RecaplyUpdated 11 Nov

Install Objects Detected in a Photo

After the download, ask your agent to add it, or run the line yourself.

Read https://recaply.co/skills/object-detection-scene/install.md and follow it to install the object-detection-scene skill from object-detection-scene.zip.

Paste it into your agent. It reads our install steps, finds the ZIP you downloaded and adds the skill for you.

New to skills? How to install a skill

Similar skills

All 735 skills

Who Objects Detected in a Photo is for

What you get

  • An MP4 scene at 16:9, 9:16, 1:1 or 4:5
  • Boxes draw around objects in a photo one by one, each with a label and a confidence score
  • Works alone or as one scene in a longer film

Good fit

  • When a vision, scanning, inventory or AI camera product shows what it recognises

Not for

  • A whole launch film in one go (use a launch film)
  • Filmed footage or people on camera

About this skill

You say

Detect "Plant 91%" and "Pot 96%" in our plant photo.

It said back

I made the clip: two boxes draw around the plant photo, labelled "Plant 91%" and "Pot 96%". The layout check found no problems, and I looked at the settled frame myself. I didn't play the rendered MP4.

  • Video: launch-scene/object-detection.mp4, 4.2 s, 1920×1080 at 30 fps, 137 KB.
  • Look: swiss, 16:9, with no logo or brand name since you didn't give me any.
  • Boxes: I measured them from photos/8-plant.jpg, in % of the picture.
  • Plant: x 16, y 20, w 65, h 49.
  • Pot: x 39, y 58, w 23, h 30.
  • Scores: only the 91% and 96% you gave me.
Show more

What it did in run 1

  1. object-detection-scene
  2. **/*.{jpg,jpeg,png,webp,JPG,PNG}
  3. ./photos/8-plant.jpg
  4. node .claude/skills/object-detection-scene/scripts/check.mjs
  5. mkdir -p launch-scene && cp photos/8-plant.jpg launch-scene/plant.jpg && cat > launch-scene/film.json <<'EOF' { "look": "swiss", "size": "16:9", "out": "object-detection.mp4", "scenes": [ { "scene": "object-boxes", "image": "plant.jpg", "boxes": [ { "x": 16, "y": 20, "w": 65, "h": 49, "label": "Plant", "score": "91%" }, { "x": 39, "y": 58, "w": 23, "h": 30, "label": "Pot", "score": "96%" } ] } ] } EOF S=.claude/skills/object-detection-scene/scripts/film.mjs node $S launch-scene/film.json --check && node $S launch-scene/film.json && ls -la launch-scene
  6. mkdir -p launch-scene && cp photos/8-plant.jpg launch-scene/plant.jpg
  7. ./launch-scene/film.json
  8. node .claude/skills/object-detection-scene/scripts/film.mjs launch-scene/film.json --check
  9. ./launch-scene/film-check.png
  10. node .claude/skills/object-detection-scene/scripts/film.mjs launch-scene/film.json && ls -la launch-scene && ffprobe -v error -show_entries format=duration -of csv=p=0 launch-scene/object-detection.mp4
  11. Done

10 Nov 2026 · Claude Code 2.1.322 with Claude Sonnet 5.5 · 33 s · $0.10 of API use

Questions

Is Objects Detected in a Photo free?

No. Objects Detected in a Photo costs $9.99 once and stays in your Recaply account, or it comes with the Recaply plan ($19.99 a month), which opens every premium skill and prompt.

Which agents can use Objects Detected in a Photo?

Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot and OpenCode.

What do I get when I buy Objects Detected in a Photo?

A ZIP of the whole skill folder (60 KB): the SKILL.md your agent follows and its 12 other files, the references, templates and scripts it uses. You download it from this page whenever you need it, and install it in Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot and OpenCode. The Recaply plan opens it too, with every other premium skill and prompt.

How do I install Objects Detected in a Photo?

Buy it, then download the ZIP from this page. For Claude Code, unzip it into .claude/skills/ in your project. For Codex, Cursor and Gemini CLI, unzip it and run npx skills add ./object-detection-scene with your agent's flag.

What do I get from Objects Detected in a Photo?

An MP4 scene at 16:9, 9:16, 1:1 or 4:5. Boxes draw around objects in a photo one by one, each with a label and a confidence score. Works alone or as one scene in a longer film.

What does Objects Detected in a Photo need?

Node 22+, Chrome and ffmpeg, and no API key. Each use costs nothing beyond your agent.

How big is Objects Detected in a Photo?

13 files, about 40,696 tokens in all. The SKILL.md is about 1,686 tokens, and your agent reads it only when the job fits.

What does a run cost?

Our 1 runs took 33 s to 33 s and $0.10 to $0.10 of API use each, in Claude Code with claude-sonnet-5-5. On a Claude plan, a run counts toward your usage instead.

Do I have to name Objects Detected in a Photo to use it?

Type /object-detection-scene first to make sure it loads. Without it, Claude Code may judge a small job simple enough to do without the skill.