Skip to content

v0.7.1

Choose a tag to compare

View all tags
@sboily sboily released this 22 May 12:52
· 93 commits to main since this release
v0.7.1
4cb17f4

What's Changed

  • chore(format): satisfy ruff format on ollama provider test
  • fix(providers/ollama): mint unique synthesized tool-call ids
  • fix(core): symmetrize BEFORE_BROADCAST block on the reentry path
  • chore(format): satisfy ruff format on inbound mixin slices
  • chore(lint): satisfy ruff after _RECENT_EVENTS_LIMIT introduction
  • feat(ai): publish THINKING_DELTA ephemerals so remote subscribers see reasoning live
  • refactor(core): raise recent_events ceiling from 50 to 2000
  • remove debug in examples
  • feat(ai): stream thinking inline via ThinkingDeltaMarker + Ollama effort levels
  • feat(providers): add native Ollama provider via ollama-python
  • fix(teams): satisfy ty for botbuilder's mistyped get_paged_members
  • feat(teams): own inbound dispatch + roster lookups end-to-end

Full Changelog: v0.7.0...v0.7.1

Assets 2
Loading