Skip to content

v0.1.18

Choose a tag to compare

View all tags
@silversurfer562 silversurfer562 released this 16 May 11:22
v0.1.18
6d95ee1
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

Two parallel tracks ship together:

Phase 1 of v1.0 roadmap — release-quality baseline + CI gate. Every PR is now gated against locked retrieval + faithfulness thresholds (precision_at_1 ≥ 0.95, recall_at_3 ≥ 1.00, mean_faithfulness ≥ 0.9686), measured empirically at mean − 2σ from N=20 back-to-back runs.

API-surface groundwork toward 0.2.0 freeze — public surface documented and snapshot-tested (tests/unit/test_api_surface.py). Formal SemVer commitments still begin at 0.2.0; the 0.1.x framing is honor-system. See docs/POLICY.md.

Highlights:

  • AttuneHelpCorpus re-exported at the package root
  • Five attune_rag.editor._* submodules renamed to drop the misleading underscore prefix; underscored paths remain as DeprecationWarning shims until 0.3.0
  • New author optional extra pinning attune-author>=0.13.0 for .help/templates/ maintenance
  • New Makefile with feature-probed help-template regen targets (sidesteps the stale-__version__ packaging bug in attune-author's 0.13.0 wheel)
  • attune-help floor bumped past yanked 0.10.0 → 0.10.1

Full changelog: https://github.com/Smart-AI-Memory/attune-rag/blob/v0.1.18/CHANGELOG.md#0118---2026-05-16

Roadmap: docs/specs/ROADMAP-v1.md
API spec: docs/specs/api-v0.2-public-surface/

Assets 2
Loading