Commit graph

4 commits

Author SHA1 Message Date
aaltshuler
16e4a833c0 Wire cluster embedding providers 2026-06-16 04:02:08 +03:00
Ragnor Comerford
ffb4a2c9ab
fix(embedding): openai-alias api key + deadline scope (PR review)
openai-alias api key (Cursor High): key resolution was dispatched on the Provider enum, which is OpenAiCompatible for both openai and openai-compatible, so OMNIGRAPH_EMBED_PROVIDER=openai (base api.openai.com) could send OPENROUTER_API_KEY and 401. Fix: the alias match now yields the ordered key-env list too, so base, model and key are all alias-derived in one place. openai takes only OPENAI_API_KEY (errors loudly if absent); openai-compatible/unset prefer OPENROUTER_API_KEY then OPENAI_API_KEY. Closes the class, not the instance.

deadline scope (Cursor Medium): the deadline bounds every embed call (query and document), which is correct, but the name said query-only. Renamed OMNIGRAPH_EMBED_QUERY_DEADLINE_MS -> OMNIGRAPH_EMBED_DEADLINE_MS (field query_deadline_ms -> deadline_ms) and updated the doc wording so the name matches the behavior. Two new alias-key tests; 20 embedding unit tests pass.

Resolved earlier in 30377c4: openai-alias base URL, single embed-model source, stale @embed ingest docs. Declined: RFC Status (the lifecycle keeps it Proposed while the PR is open).
2026-06-15 21:44:31 +02:00
Ragnor Comerford
70ed848b9d
docs(embeddings): @embed model arg + same-space validation (RFC-012 Phase 3-4)
Document the optional @embed model kwarg, the query-time same-space rejection, model-string strictness, and the loud schema-apply refusal on model change. Mark RFC-012 phases 1-4 implemented.
2026-06-15 21:09:35 +02:00
Ragnor Comerford
5b83e6b56d
docs(rfc): add RFC-012 provider-independent embedding configuration
Records the design for one resolved EmbeddingConfig behind a sealed Provider enum (OpenAiCompatible primary, covering OpenRouter as the default gateway, OpenAI direct, and self-hosted endpoints; native Gemini for task-type asymmetry; Mock). Identity recorded in the schema IR with query-time same-space validation, NFR floor (deadline, observability, single normalize), and a 5-phase rollout. Corrects the prior handoff's misdiagnosis (the OpenAI client it blamed was dead code).

Links from docs/dev/index.md.
2026-06-15 15:07:38 +02:00