clawdie-ai/scripts/memory
Clawdie AI e73afafbb2 feat: switch embeddings to OpenAI text-embedding-3-small; add embed-docs.py
- embed-docs.py: new script to batch-embed project docs into klavdija_brain
  (37 docs → 852 chunks with session_id pattern doc:<path>)
- import-memories.py: updated embed() to send Authorization header and
  dimensions=1024 parameter for OpenAI/configurable provider support
- memories-transfer-v2.json: recovered historical session memories (local copy)
- package.json: add @supabase/supabase-js dependency
- src/channels/telegram.ts, src/config.ts: Telegram channel integration

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

---
Build: pass | Tests: pass — Tests  431 passed (431)
2026-03-28 20:44:09 +00:00
..
chunk.sh Add memory pipeline scripts (Phase 2) 2026-03-10 09:30:34 +00:00
common.sh refactor: derive DB identity from AGENT_NAME, drop hardcoded agent refs 2026-03-28 08:55:06 +00:00
embed-docs.py feat: switch embeddings to OpenAI text-embedding-3-small; add embed-docs.py 2026-03-28 20:44:09 +00:00
embed.sh refactor: derive DB identity from AGENT_NAME, drop hardcoded agent refs 2026-03-28 08:55:06 +00:00
import-memories.py feat: switch embeddings to OpenAI text-embedding-3-small; add embed-docs.py 2026-03-28 20:44:09 +00:00
memories-transfer-v2.json feat: switch embeddings to OpenAI text-embedding-3-small; add embed-docs.py 2026-03-28 20:44:09 +00:00
memory-hydrate-pg.sh refactor: derive DB identity from AGENT_NAME, drop hardcoded agent refs 2026-03-28 08:55:06 +00:00
memory-pg.sh fix(dates): standardize DD.mmm.YYYY display and snapshot naming 2026-03-14 00:07:55 +01:00