SpecStream for pi-websearch
SpecStream is the live web data engine for teams and agents. One websearch.json file points pi-websearch's tavily backend at SpecStream, and the pi agent searches the live web at $1 per 1,000 searches of up to 10 results.
What you get
- No code change
- pi-websearch keeps its tavily backend. Only baseUrl and apiKey change.
- Domain filters work
- allowed_domains and blocked_domains reach SpecStream as include_domains and exclude_domains, and both apply.
- A free backup
- List DuckDuckGo HTML after SpecStream, and pi-websearch moves to it when a search fails.
Set it up
- Run pi install git:github.com/code-yeongyu/pi-websearch.
- Write ~/.pi/websearch.json with backend tavily, your SpecStream key as apiKey, baseUrl https://api.specstream.ai/tavily/search and auto false.
- Run /websearch status in pi, then ask about something from this week.
The full setup guide has every step, with what to do if something doesn’t arrive.
Live web data, in pi-websearch.
Ask in your own words for companies, places, people, filings or anything else the web lists.