Skip to content

Add exact one-step policy sensitivity evidence - #8

Merged
omar07ibrahim merged 17 commits into
mainfrom
portfolio/policy-sensitivity-evidence
Aug 9, 2026
Merged

Add exact one-step policy sensitivity evidence#8
omar07ibrahim merged 17 commits into
mainfrom
portfolio/policy-sensitivity-evidence

Conversation

@omar07ibrahim

@omar07ibrahim omar07ibrahim commented Aug 9, 2026

Copy link
Copy Markdown
Owner

Outcome

Adds an exact, entropy-free one-step policy-minimum sensitivity analysis to the Python API and CLI, with text, JSON, and CSV contracts plus source-bound visual evidence.

What is reviewable

  • exact baseline and one-class-at-a-time relaxed counts from bounded integer dynamic programming;
  • independently checkable reduced baseline/relaxed fractions and policy fingerprints;
  • explicit boundary: one step only, effects are not additive, no candidate, no entropy;
  • a pure fail-closed SVG renderer and a real deterministic CLI transcript renderer;
  • generator, independent evidence checker, distribution attestation, README workflow, and negative-path tests.

Evidence contract

  • 27 manifest-recorded artifacts are bound to 39 source files on final head 36aab7fd69f6f59588e75278cfe0d475389424ed.
  • New evidence: policy-sensitivity.json, policy-sensitivity.txt, a 1600×1366 real CLI capture, and an 1800×850 exact-impact SVG.
  • The hosted candidate from Verify run 31325386816 passed recapture and fail-closed drift inventory before byte-for-byte adoption.
  • Independent ZIP validation checked every path, regular-file mode, byte count, SHA-256, MIME, image geometry, canonical manifest encoding, source binding, and common secret markers.
  • The README embeds 15 real/reproducible visuals with non-empty alt text and valid repository paths.

Verification

  • Python 3.11, 3.12, 3.13, and 3.14: 316 tests per runtime.
  • 595 statements and 144 branches at 100% coverage; Ruff, formatting, and strict mypy clean.
  • Canonical distribution rebuild and installed-wheel smoke passed.
  • Final browser replay was byte-identical with a clean checkout: Verify run 31326047648.
  • CodeQL Analyze (actions) and Analyze (python) passed: run 31326045169.
  • Independent inclusion–exclusion recomputation matched the baseline, all four relaxed totals/additions, and every reduced fraction.
  • Open code-scanning, Dependabot, and secret-scanning alerts: 0 / 0 / 0.
  • 17 linear commits, behind=0; every author and committer is Omar Ibrahim, with no co-authors.

No password is sampled by this analysis or its evidence pipeline.

Post-main verification

  • main is the exact reviewed head 36aab7fd69f6f59588e75278cfe0d475389424ed; PR Add exact one-step policy sensitivity evidence #8 closed as a fast-forward with no merge commit.
  • Post-main Verify run 31326474561 passed all six jobs, including byte-identical Chromium recapture.
  • Post-main CodeQL run 31326474180 passed Analyze (actions) and Analyze (python).
  • main requires those eight exact checks with strict mode, admin enforcement, linear history, resolved conversations, and no force-push or deletion.
  • Open code-scanning, Dependabot, and secret-scanning alerts remain 0 / 0 / 0.

@omar07ibrahim
omar07ibrahim marked this pull request as ready for review August 9, 2026 17:27
@omar07ibrahim
omar07ibrahim merged commit 36aab7f into main Aug 9, 2026
9 checks passed
@omar07ibrahim
omar07ibrahim deleted the portfolio/policy-sensitivity-evidence branch August 9, 2026 17:28
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant