Image Well

Search and download images from 15 sources through a single CLI. Seven sources work with zero API keys (Openverse, Wikimedia, Met Museum, Cleveland Museum, Art Institute of Chicago, Getty Museum, NASA). Additional sources activate when keys are set.

Quick Start

# Search across all no-key sources
uv run ~/.claude/skills/image-well/scripts/well.py search "ancient Minoan fresco"

# Use a preset for domain-specific searches
uv run ~/.claude/skills/image-well/scripts/well.py search "F-35 fighter jet" --preset military

# Download results with metadata sidecars
uv run ~/.claude/skills/image-well/scripts/well.py search "sunset" --format download --output ./images/

# Check which sources are available
uv run ~/.claude/skills/image-well/scripts/well.py sources

# Visual HTML preview — opens in browser, no context pollution
uv run ~/.claude/skills/image-well/scripts/well.py search "bronze statue" --format html

# Output as JSON for piping
uv run ~/.claude/skills/image-well/scripts/well.py search "cat" --format json

Sources

Add Tier 2 keys to ~/.config/env/secrets.env for expanded coverage.

Source notes:

• Cleveland downloads the ~3400px print JPEG; the ~900px web derivative is the thumbnail. The archival TIFFs are never fetched.
• Art Institute of Chicago images are public domain (PD), not formally CC0 — --license cc0 accepts them. Downloads use AIC's preferred 843px IIIF size and are serialized (1/s) per their API etiquette.
• Getty rides the collection website's undocumented JSON search API (open_content=true filter) — the only keyword-search surface Getty exposes. Every emitted row is gated on the CC0 URI in its IIIF manifest license; downloads at 1200px IIIF. Superb Greek/Roman/Etruscan; thin on Egypt/Near East.

Museum Collections on Wikimedia Commons

Some museums have no usable API but mass-uploaded their images to Wikimedia Commons. Search them through the existing wikimedia source with a category-scoped query:

The Walters v1 API closed in 2023 and the whole thewalters.org domain (including its image CDN) sits behind Cloudflare bot protection — Commons is the only programmatic path. Note the 2012 Walters upload was recorded as CC BY-SA 3.0 (many files also carry a Public Domain Mark); keep the per-file license Commons reports rather than assuming CC0.

Presets

Options

--sources NAME [...]    Specific sources to search
--preset NAME           Use a preset group
--limit N               Max results per source (default: 10)
--license LICENSE       Filter: cc0, cc-by, cc-by-sa, any
--format FORMAT         table, json, download, urls, html, tunnel
--output DIR            Download directory (default: ./well-images/)
--no-cache              Skip result cache
--timeout N             Per-source timeout in seconds (default: 15)
--quiet                 Suppress progress output

Workflows

Source images for a project

# Search, review in table, then download the good ones
uv run well.py search "Mediterranean landscape" --preset stock --format json > results.json
# Review results.json, then download specific URLs

Build an image corpus

# Download CC0 images for a research project
uv run well.py search "Bronze Age pottery" --preset museum --license cc0 --format download --output ./corpus/

Pipe to image-forge for processing

# Search, download, then resize with image-forge
uv run well.py search "sunset" --format download --output ./raw/
# Then use image-forge skill to process downloaded images

Cache

Results are cached for 24 hours at ~/.cache/image-well/.

uv run well.py cache stats    # Show cache size
uv run well.py cache clear    # Clear all cached results