Agent Evidence SE Workflow Case Prototype v0.1.0-rc3
Pre-release
Pre-release
·
77 commits
to main
since this release
Agent Evidence SE Workflow Case Prototype v0.1.0-rc3
This prerelease candidate repairs rc2 release-asset reproducibility.
What changed from rc2
- Adds top-level evidence documents: CLAIM_BOUNDARY.md, ENVIRONMENT.md, REPRODUCIBILITY.md, ARCHIVE_NOTES.md, and AI_DISCLOSURE.md.
- Corrects stale unrelated citation metadata to the SE workflow case prototype line.
- Packages an installable standalone source tree with pyproject.toml and the minimal agent_evidence module required by the SE case runners.
- Reports the selected pytest suite truth from rc3: 28 passed.
Evidence summary
- 18 total rows: 3 valid, 15 invalid.
- Diagnostic match: 18/18.
- Schema-only missed invalid: 15.
- Log-only missed invalid: 15.
- Policy-only missed invalid: 13.
Claim boundary
Representative reproducible SE workflow fixtures only. No industrial real-world evaluation, production deployment, benchmark superiority, legal/compliance sufficiency, full FDO interoperability, or TSE v3 readiness claim.