Skip to content

Cache Tavily responses across forecasting workers - #46

Merged
exe-dev-github-integration[bot] merged 1 commit into
mainfrom
feat/tavily-response-cache
Oct 27, 2026
Merged

exe-dev-github-integration[bot] merged 1 commit into
mainfrom
feat/tavily-response-cache

Conversation

@exe-dev-github-integration

Copy link
Copy Markdown
Contributor

Summary

  • Add shared, one-hour disk caching for Tavily search and full-page extraction in both parity-tested tool copies.
  • Coalesce concurrent identical requests; preserve per-caller tags and current URL guards. Do not cache failures; fail open on storage errors.
  • Add coverage for parameters, freshness, account partitioning, pagination, errors, guards, and concurrent processes, plus an opt-in budgeted live probe.

Verification

  • 46 relevant unit tests passed during development.
  • Live probe: 24 tool calls from 8 concurrent workers across both implementations resulted in exactly 1 Tavily extraction and 1 search. No model calls.
  • No forecast data or schedules changed.

Deployment

Normal airo deploy after required checks and merge, retaining the published forecast snapshot.

@exe-dev-github-integration
exe-dev-github-integration Bot merged commit febd376 into main Oct 27, 2026
4 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

0 participants