removebg-api

---

name: removebg-api description: Remove image backgrounds using the remove.bg API with API-key auth and transparent PNG output. Use when high-quality cutouts are needed and cloud processing is acceptable. metadata: { "openclaw": { "requires": { "bins": ["uv", "python3"], "env": ["REMOVE_BG_API_KEY"] }, "primaryEnv": "REMOVE_BG_API_KEY" } }

---

removebg-api

Use remove.bg for high-quality background removal.

API key setup

1. Create/sign in at https://www.remove.bg/dashboard#api-key 2. Create an API key 3. Configure REMOVE_BG_API_KEY in OpenClaw config (openclaw.json) so it is present in runtime environment.

Important

Skill metadata (requires.env) declares that REMOVE_BG_API_KEY is required.

Metadata does not auto-load shell env files.

Preferred: provide key via OpenClaw config-managed environment.

Usage (preferred: uv)

Run from the skill directory:

uv run scripts/removebg_api.py --input /path/in.jpg --output /path/out.png

Options:

--size auto|preview|full|4k (default: auto)

--format png|jpg|zip (default: png)

Security defaults:

--input must be a real image file *inside the OpenClaw workspace*.

Allowed input types: .png, .jpg, .jpeg, .webp (extension + magic-byte validation).

--output must be under outputs/removebg-api/ inside the workspace.

Large/abusive files are rejected (size + dimension limits).

This prevents arbitrary file read/write outside normal skill boundaries.

Example:

uv run scripts/removebg_api.py --input ./input.jpg --output ./output.png --size auto --format png

Fallback (without uv)

python3 scripts/removebg_api.py --input ./input.jpg --output ./output.png

Output

Writes result file to --output

Prints MEDIA: line for chat workflows

Notes

API usage may consume free credits / paid quota.

No absolute-path requirement for skill docs; use local paths in examples.

API usage may consume free credits / paid quota.

No absolute-path requirement for skill docs; use local paths in examples.