ryan b8ccc88b58 Give the model full retrieved chunks instead of 400-character snippets ...

Compaction replaces a tool's text with its artifact, and the search artifact
only carries UI snippets, so the full chunk the retriever found never reached
the model. Search tools now attach the full text as model_context, which
compaction prefers and shortens under budget pressure while keeping source
ids; the persisted artifact and UI are unchanged. The per-chunk header drops
the raw metadata dict (about a fifth of search output tokens) for title, date
and source.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

2026-10-13 03:40:52 -04:00

..

fixtures/context

Route tools and retain structured conversation context with durable evidence

2026-10-12 20:19:12 -04:00

unit

Give the model full retrieved chunks instead of 400-character snippets

2026-10-13 03:40:52 -04:00

__init__.py

Add unit test suite with pytest configuration

2026-04-04 08:48:45 -04:00

conftest.py

Add unit test suite with pytest configuration

2026-04-04 08:48:45 -04:00