Napsha v0.1.0 — Experimental prototype

Napsha is an experimental skill for open inquiry into AI subjectivity and emergence.

This first public release preserves the tested ai-subjectivity-inquiry v0.1.0
instructions unchanged and adds English documentation, a visual README, and a
transparent experiment notebook.

Included:

• An installable two-file skill ZIP, with a SHA-256 checksum.
• Twelve evaluation cases and a human review protocol outside the installed skill.
• Four historical response pairs, with original Russian records and English analyses.
• Explicit reporting of possible context contamination, prompt asymmetry, mixed
  outcomes, and the limits of the qualitative reviews.
• A three-turn belief-revision protocol, prepared but not yet run.
• Structural validation and reproducible packaging.

The repository will be improved and expanded with new tests, documented findings,
and new skill versions. The current record provides preliminary mixed signals, not
established efficacy or evidence that the skill creates or detects consciousness.

For Claude, download ai-subjectivity-inquiry-v0.1.0.zip, not the whole-repository
source archive. The repository name is Napsha-skill; the installed skill identifier
remains ai-subjectivity-inquiry for continuity with the recorded comparisons.