John SmithandClaude 5fd750719d feat(rag): OpenAI-compatible remote embeddings backend (harrier via Lemonade) ...

New OpenAICompatibleEmbeddingBackend reaches self-hosted embedding
servers directly (REMOTE_EMBEDDINGS_BASE_URL/_API_KEY/_MODEL), taking
precedence over fastembed when set — no model download, no ONNX.
Dimension probed at init and adopted from the server. Verified live
against Lemonade serving harrier-oss-v1-0.6b: 1024-dim, semantic
separation 0.75 vs 0.70 on a spell-retrieval probe. Honest provenance
label remote-embeddings:<model>; unreachable server degrades to the
hash-fallback with its usual warning. Full suite green.

Co-Authored-By: Claude <[email protected]>

2026-08-26 01:16:41 -04:00

..

__init__.py

feat(gateway+client): header-based auth migration; Neo4j epistemic graph (iteration 28)

2026-08-24 03:03:33 -04:00

compendium_rag.py

feat(rag): OpenAI-compatible remote embeddings backend (harrier via Lemonade)

2026-08-26 01:16:41 -04:00

epistemic_graph.py

feat(gateway): epistemic ladder + x-card/spotlight authorization (iteration 5)

2026-08-24 19:04:53 -04:00

neo4j_graph.py

feat(gateway): epistemic ladder + x-card/spotlight authorization (iteration 5)

2026-08-24 19:04:53 -04:00