Michael Chihlas 2007dcb990 feat: add Anthropic prompt caching to assistant chat ...

Cache the static system prompt and conversation history prefix across
turns, reducing input token costs by ~80% on multi-turn conversations.
RAG context is intentionally uncached since it changes per query.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

2026-03-05 18:17:55 -05:00

..

api

fix: RAG vector search SQL syntax error breaking assistant chat

2026-03-05 11:41:22 -05:00

core

feat: add survey invite email template

2026-03-05 01:52:47 -05:00

models

feat: add survey_invites migration and invite_id FK

2026-03-05 01:41:22 -05:00

schemas

feat: admin survey responses page with expandable detail and CSV export

2026-03-05 07:55:49 -05:00

services

feat: add Anthropic prompt caching to assistant chat

2026-03-05 18:17:55 -05:00

__init__.py

Initial commit: Backend API Phase 1a complete

2026-01-22 14:38:53 -05:00

main.py

feat: Slate & Ice Modern aesthetic redesign (#94)

2026-03-04 20:44:25 -05:00