Skip to content

v0.1.4

Latest
Latest

Choose a tag to compare

View all tags
@github-actions github-actions released this 25 Aug 15:19
v0.1.4
cc3826f
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

0.1.4 (2026-08-25)

Features

  • db: memory_labels multi-label join table (m2.5.G5.M2) (#586) (28cdfe0)
  • db: temporal validity foundation on memories (m2.5.G5.M1) (#581) (8bf93e3)
  • embedder: add tei gpu embedding backend (#561) (04f24f5)
  • embedder: hosted-API embedding backend (OpenAI default, Cohere alternate) (m2.5.G4.M3) (#576) (671212b)
  • embedder: interface registry and Python ABC (m2.5.G4.M1) (#557) (640afed)
  • embedder: redis content-hash embedding cache (m2.5.G4.M4) (#579) (a395502)
  • mcp: mcp server skeleton with stub tools and auth boundary (m2.5.G6.M1) (#590) (50d161a)
  • obs: phase 2.5 exit gate with local LGTM stack (m2.5.G7) (#592) (e66e359)
  • proxy: add model capability registry for per-model context and features (m2.5.G1.M2) (#541) (3eb3ee5)
  • proxy: anthropic provider adapter (2.5.G1.M1) (#538) (65a6360)
  • proxy: non-streaming response pipeline seam (m2.5.G3.M1) (#551) (60c23b8)
  • proxy: openaicompatible adapter for self-hosted LLM backends (m2.5.G1.M3) (#544) (033cb62)
  • schema: memory_relationships graph foundation (m2.5.G5.M3) (#588) (80c8dae)
  • tokenizer: per-family token counting registry (m2.5.G2.M1) (#548) (6a08719)
Assets 2
Loading