Fetch full page content per result, then call Gemini 2.5 Flash (via OCI GenAI) to extract: abstract, date, author, tags, content_type, and uuid. Enrichment runs in parallel (ThreadPoolExecutor). enrich=False flag available to skip for raw/fast results.
7 lines
270 B
Plaintext
7 lines
270 B
Plaintext
SEARXNG_URL=https://searxng.cloud-handson.com
|
|
|
|
# OCI GenAI (auth via ~/.oci/config)
|
|
OCI_COMPARTMENT_ID=
|
|
OCI_GENAI_ENDPOINT= # e.g. https://inference.generativeai.us-ashburn-1.oci.oraclecloud.com
|
|
OCI_CHAT_MODEL_ID= # OCID of the model (use google.gemini-2.5-pro OCID)
|