skill

stock-fetch

Search & download free-license stock media (images/videos/music/SFX) from Pixabay & Mixkit into a type-based asset library. Pure stdlib, agent-skill ready.

ghgjkbf/stock-fetchcommunitylistedunreviewed

Use skill

Set up

Skills are installed by placing their folder where your agent loads skills.

Download ZIP from GitHub Open the repository

The download comes from github.com/ghgjkbf/stock-fetch itself, not from our server, so what you get is exactly what its author published.

Install instructions come from the README

Skills do not use an MCP config. Follow the steps in the README below.

README

synced Oct 22, 2026

stock-fetch

English | 中文

Search and download free-license stock media from the command line — images, videos, music, and sound effects — into a type-based local asset library.

Built for AI-agent workflows: when a task needs real-world footage or assets (not AI-generated ones), the agent runs this script, gets JSON back, and the files are already filed under the right folder.

Why

Free stock sites are scattered and network-reachable from some machines but not others. stock-fetch is deliberately small:

Two working sources, one command. Pixabay (official API, free key) covers images + videos; Mixkit (no key, page parsing) covers videos + music + SFX.
Type-based filing. Downloads land in assets/images | video | bgm | audio automatically, so an asset library stays organized without manual moves.
License info in the output. Every result carries its license string; agents can cite it without a second lookup.
Zero dependencies. Pure Python standard library, JSON on stdout, exit codes that behave.

Quick start

# search only
python scripts/stock_fetch.py search -q "sunset ocean" --kind image -n 5

# download top 2 matching videos
python scripts/stock_fetch.py get -q "city timelapse" --kind video -n 2

# music and sound effects (no key needed)
python scripts/stock_fetch.py get -q "upbeat corporate" --kind music -n 3
python scripts/stock_fetch.py get -q "click" --kind sfx -n 3

Result JSON:

{
  "ok": true,
  "downloaded": 1,
  "failed": 0,
  "items": [
    {
      "source": "mixkit",
      "type": "video",
      "url": "https://assets.mixkit.co/videos/41576/41576-1080.mp4",
      "license": "Mixkit License (free for commercial use, no attribution)",
      "saved": "assets/video/mixkit_sunset_80676.mp4",
      "bytes": 187175135
    }
  ]
}

Configuration

Setting	How
Pixabay key	free key from pixabay.com/api/docs — put `PIXABAY_API_KEY=xxx` in `scripts/.env`, or export the env var
Asset library root	`STOCK_FETCH_ROOT` env var, else `./assets` relative to the working directory
Env file location	`STOCK_FETCH_ENV` env var, else `scripts/.env` next to the script
Destination override	`--dest video/city` (relative to the library root)

Rate limits: Pixabay allows 100 requests / 60 seconds and forbids systematic bulk downloading — this tool is for picking a handful of assets per task, not mirroring the catalog.

Sources

Source	Media	Key	Notes
Pixabay	images + videos	free key	official API, safe-search on
Mixkit	videos + music + SFX	none	search-page parsing; 1080p links inferred from the fixed naming pattern
Wikimedia Commons	CC / public-domain images	none	fallback via `--source wikimedia`; keep attribution

Use as an agent skill

SKILL.md is included, so npx skills add <this repo> (or any skill loader that reads SKILL.md) can expose it to your agent directly.

License

Apache-2.0