Markus 8cbd55c963 feat(p2): embedding cache in the Embedder base, fused hybrid search, rebuild paths
Caching/batching/prefixing moved into Embedder so FakeEmbedder exercises the
same cache the Bifrost client uses. Rebuilding an index with an unchanged model
embeds nothing; a dim or model change drops the vectors and re-embeds. A
token-free query short-circuits before the embedding call.
2026-09-12 07:56:06 +00:00
2026-09-12 07:39:27 +00:00
2026-09-12 07:39:27 +00:00

mdmem

Markdown-canonical agent memory service. See PLAN.md for the full design.

S
Description
No description provided
Readme
541 KiB
0 Stars 1 Watchers 0 Forks
Languages
Python 92.4%
HTML 3.7%
CSS 3.6%
Dockerfile 0.3%