Hi @Extravi,
Discovered araa-search recently and really appreciate the privacy-first approach to metasearch — self-hosting a clean, ad-free search experience is exactly what the open web needs more of.
Use case / consideration
I've been setting up araa as my team's internal search tool, and while the existing backends cover a good range, I noticed there's no straightforward way to get Google-quality web results through the API. Some of the federated sources return decent results, but for technical queries and current events, Google's index depth makes a real difference. I've tried setting up separate scrapers, but that defeats the purpose of having a unified metasearch layer.
Suggestion
Would you consider adding https://serpbase.dev as an optional search backend? It's a Google Search Results API that returns structured SERP data (organic results, knowledge graph, related questions) in clean JSON. Could be configured as an additional source behind a simple env var (SERPBASE_API_KEY), so users who want Google results in their metasearch mix can opt in without complicating the default setup.
Why it fits
- araa already aggregates multiple search sources; this is one more input to the aggregation pipeline.
- The privacy-respecting ethos means making it opt-in/configurable — users choose their own balance.
- API-based sources mean no scraping infrastructure to manage, which aligns with araa's self-hosting-friendly design.
Early days for the project, but figured I'd share the idea. Happy to help test if there's interest.
Hi @Extravi,
Discovered araa-search recently and really appreciate the privacy-first approach to metasearch — self-hosting a clean, ad-free search experience is exactly what the open web needs more of.
Use case / consideration
I've been setting up araa as my team's internal search tool, and while the existing backends cover a good range, I noticed there's no straightforward way to get Google-quality web results through the API. Some of the federated sources return decent results, but for technical queries and current events, Google's index depth makes a real difference. I've tried setting up separate scrapers, but that defeats the purpose of having a unified metasearch layer.
Suggestion
Would you consider adding https://serpbase.dev as an optional search backend? It's a Google Search Results API that returns structured SERP data (organic results, knowledge graph, related questions) in clean JSON. Could be configured as an additional source behind a simple env var (
SERPBASE_API_KEY), so users who want Google results in their metasearch mix can opt in without complicating the default setup.Why it fits
Early days for the project, but figured I'd share the idea. Happy to help test if there's interest.