Google Images Search
Search Google Images through Frevana.
Purpose
This skill calls Frevana's /service/google-images endpoint and returns the SerpAPI response unchanged.
Required input: q.
Supported request fields:
q(required)locationuulegoogle_domainglhlcrdevicestartijntbschipsimgarimgszsafenfprfilterimage_sizeimage_typeimage_colorlicensesperiod_unitperiod_valuestart_dateend_dateno_cacheasynczero_trace
The script validates that the response is JSON and prints it to stdout. Do not rewrite or reshape returned data unless the user explicitly asks for a transformation.
What This Skill Needs
- query
FREVANA_TOKENin the environment variables, or--tokenfor the current runcurl,bash, andpython3
The API base URL defaults to https://ai-factory.frevana.com. Set FREVANA_API_BASE_URL to point at another host.
Commands
Search images
bash <skill-path>/scripts/search_google_images.sh \
--q "coffee shop interior" \
--gl "us" \
--hl "en"
Filter images
bash <skill-path>/scripts/search_google_images.sh \
--q "coffee shop interior" \
--image-color "blue" \
--safe "active"
Save JSON to a file
bash <skill-path>/scripts/search_google_images.sh \
--q "Q" \
--output "./out/google-images-search.json"
Fixed Request Shape
The script sends this payload shape, omitting optional fields that were not provided:
{
"q": "Q",
"location": "LOCATION",
"uule": "UULE",
"google_domain": "GOOGLE_DOMAIN",
"gl": "GL",
"hl": "HL",
"cr": "CR",
"device": "DEVICE"
}
Do not pass unsupported fields. Do not pass SerpAPI engine or api_key; Frevana sets those server-side. Use --output for local file saving.
Output
- Success: validated JSON on stdout
- With
--output: the same JSON is also written to the requested file - Failure: response body or parsing error on stderr, with a non-zero exit code
Example Prompts
- "Search Google Images for query."
- "Run Google Images Search and save the raw JSON."