Skip to content

kernel_chat 0.8.0

Choose a tag to compare

View all tags
@GrazianoGuiducci GrazianoGuiducci released this 20 Sep 20:46
Immutable release. Only release title and notes can be modified.
v0.8.0
e5520de
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

kernel_chat 0.8.0

kernel_chat 0.8.0 deepens FDLA with consequence-aware temporal recomposition
while preserving the existing in-flow correction relation.

Main change

A later consequence or new evidence can make a relation understandable that was
not available in the earlier field. FDLA now preserves these as distinct when
material:

what happened
!= what was understood then
!= what is understood now

The current understanding is not narrated backward as earlier knowledge, and an
earlier faithful interpretation is not frozen after a material consequence
changes what can now be understood.

AGENTS exposes the later-consequence condition to the FDLA owner. Structural
validation and the independent rendered Markdown consumer proof protect that
route, while README and Architecture expose the broadened current capability.

Compatibility

  • existing in-flow FDLA remains valid;
  • portable entry remains 3.0.0;
  • instance schema remains kernel_chat.instance.v1;
  • configurator, ChatGPT adapter, state and operations contracts are unchanged;
  • no configured bridge, installed host instruction or user-instance migration
    follows automatically from this release.

Material review

Reviewed semantic candidate:
ac74126f0cad31b6926e72b184df3ad0104bcf20.

  • CI 35533925195: success;
  • validator: valid=true, errors=[], warnings=[];
  • tests: 64/64;
  • matrix: 8/8 Python 3.11–3.14 × Ubuntu/Windows;
  • review result: no_material_blocker.

The 0.8.0 promotion commit must receive its own green CI before protected tag
and immutable release publication.

Assets 3
Loading