PyPI - falsify - Versions diffs - 0.3.1__tar.gz → 0.3.2__tar.gz - Mend

falsify 0.3.1tar.gz → 0.3.2tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (85) hide show

{falsify-0.3.1/falsify.egg-info → falsify-0.3.2}/PKG-INFO RENAMED Viewed

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: falsify
-Version: 0.3.1
+Version: 0.3.2
 Summary: PRML reference CLI — pre-register an ML evaluation claim as a SHA-256 manifest; verify PASS/FAIL/TAMPERED.
 Author: Cüneyt Öztürk
 License: MIT
@@ -52,14 +52,14 @@ TAMPERED  (exit 3)
 No install? Verify any manifest in-browser at [registry.falsify.dev](https://registry.falsify.dev). Byte-equivalent reference CLIs also ship for JS (`npm i -g falsify-js`), Go, and Rust.
-4 reference implementations (Python, JavaScript, Go, Rust) byte-equivalent on all 20 conformance vectors (12 v0.1 stable + 8 v0.2). PRML v0.2 frozen 2026-05-22. The same day, Lock #2 (a public hypothesis on the spec's own distribution, target ≥3 external contributors in 14 days) resolved at 0/3. The mechanism worked, the post-mortem is at [falsify.dev/notes/lock-2-postmortem](https://falsify.dev/notes/lock-2-postmortem/). Designed for ML eval rigor. Maps to EU AI Act Article 12 evidence as a side effect.
+4 reference implementations (Python, JavaScript, Go, Rust) byte-equivalent on all 21 conformance vectors (13 v0.1 stable + 8 v0.2). PRML v0.2 frozen 2026-05-22. The same day, Lock #2 (a public hypothesis on the spec's own distribution, target ≥3 external contributors in 14 days) resolved at 0/3. The mechanism worked, the post-mortem is at [falsify.dev/notes/lock-2-postmortem](https://falsify.dev/notes/lock-2-postmortem/). Designed for ML eval rigor. Maps to EU AI Act Article 12 evidence as a side effect.
 > **Pre-registration + CI for AI-agent claims.** Lock the claim and threshold with SHA-256 *before* running the experiment — or the result doesn't count.
 ![CI](https://github.com/studio-11-co/falsify/actions/workflows/falsify.yml/badge.svg)
 ![Multi-lang Conformance](https://github.com/studio-11-co/falsify/actions/workflows/multi-lang-conformance.yml/badge.svg)
 ![PyPI](https://img.shields.io/pypi/v/falsify?color=brightgreen&label=pypi)
-![coverage](https://img.shields.io/badge/tests-564%20passing-brightgreen)
+![coverage](https://img.shields.io/badge/tests-561%20passing-brightgreen)
 ![impls](https://img.shields.io/badge/reference%20impls-4%20(py%20%C2%B7%20js%20%C2%B7%20go%20%C2%B7%20rs)-brightgreen)
 ![honesty](https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/studio-11-co/falsify/main/.falsify/badge.json)
 ![python](https://img.shields.io/badge/python-3.11%2B-blue)
@@ -148,12 +148,12 @@ This repository is the home of **PRML v0.1** — Pre-Registered ML Manifest Spec
 **Audit & compliance crosswalks** (subcategory-by-subcategory maps from major AI governance frameworks to PRML fields, FULL/PARTIAL/NONE tagged):
-- **[EU AI Act Article 12](https://spec.falsify.dev/eu-ai-act/article-12/)** — code-level pattern for the 2 August 2026 high-risk applicability deadline
+- **[EU AI Act Article 12](https://spec.falsify.dev/eu-ai-act/article-12/)** — code-level pattern for the 2 December 2027 high-risk applicability deadline
 - **[Article 12 readiness diagnostic](https://spec.falsify.dev/article-12-readiness/)** — 10-question browser-only self-assessment
 - **[NIST AI RMF 1.0 crosswalk](https://spec.falsify.dev/nist-ai-rmf/)** — GOVERN / MAP / MEASURE / MANAGE subcategory map (incl. AI 600-1 GenAI Profile)
 - **[ISO/IEC 42001:2023 crosswalk](https://spec.falsify.dev/iso-42001/)** — AIMS clause-by-clause evidence map (Clauses 7-9 + Annex A controls)
-**Long-form working notes** (2026-05-23, written for compliance leads, AI governance officers, and notified body assessors preparing for the 2 August 2026 deadline; CC BY 4.0):
+**Long-form working notes** (2026-05-23, written for compliance leads, AI governance officers, and notified body assessors preparing for the 2 December 2027 deadline; CC BY 4.0):
 - **[EU AI Act readiness assessment](https://falsify.dev/eu-ai-act-readiness/)** — six binding articles, ten-question gap check, evidence shape per obligation
 - **[2 August 2026 deadline](https://falsify.dev/ai-act-deadline-august-2026/)** — three application dates, Article 99 penalty structure, ten-week plan
@@ -162,7 +162,7 @@ This repository is the home of **PRML v0.1** — Pre-Registered ML Manifest Spec
 - **[ISO/IEC 42001 readiness](https://falsify.dev/iso-42001-readiness/)** — seven clauses, EU AI Act Article 17 overlap, twelve-month certification path
 - **[Lock #2 post-mortem](https://falsify.dev/notes/lock-2-postmortem/)** — field report on running a falsifiable spec in public
-**Reference implementations** (four languages, 12 v0.1 + 8 v0.2 candidate vectors = 20 total; multi-lang CI runs all 20 byte-for-byte per push and daily at 04:00 UTC):
+**Reference implementations** (four languages, 13 v0.1 + 8 v0.2 candidate vectors = 21 total; multi-lang CI runs all 21 byte-for-byte per push and daily at 04:00 UTC):
 - **Python:** [`falsify.py`](falsify.py) — original reference, uses PyYAML
 - **Node.js:** [`impl/js/`](impl/js/) — second reference, ~400 LOC, hand-rolled, zero deps
@@ -173,7 +173,7 @@ Hosted spec at [spec.falsify.dev/v0.1](https://spec.falsify.dev/v0.1). Public re
 **Companion projects** (separate repos under `studio-11-co`, each MIT or CC0 licensed):
-- **[`falsify-cookbook`](https://github.com/studio-11-co/falsify-cookbook)** — field manual for the spec: 11 patterns + 4 anti-patterns, every one a single page with a runnable example, including [Pattern 11: PRML + Sigstore for execution integrity](https://github.com/studio-11-co/falsify-cookbook/blob/main/patterns/11-sigstore-execution.md) closing the §8.1 gap. CC0.
+- **[`falsify-cookbook`](https://github.com/studio-11-co/falsify-cookbook)** — field manual for the spec: 13 patterns + 4 anti-patterns, every one a single page with a runnable example, including [Pattern 11: PRML + Sigstore for execution integrity](https://github.com/studio-11-co/falsify-cookbook/blob/main/patterns/11-sigstore-execution.md) closing the §8.1 gap. CC0.
 - **[`falsify-integrity-index`](https://github.com/studio-11-co/falsify-integrity-index)** — public scorecard of how 25+ well-known ML eval claims meet the 9 PRML falsifiability criteria. Live at [falsify.dev/integrity](https://falsify.dev/integrity). CC0 data, MIT tooling.
 - **[`falsify-inspect`](https://github.com/studio-11-co/falsify-inspect)** — Inspect AI adapter: anchor an Inspect AI eval claim's threshold to a SHA-256 hash before the run, verify the post-run log against it. MIT.
 - **[`prml-verify-action`](https://github.com/studio-11-co/prml-verify-action)** — composite GitHub Action ([listed on Marketplace](https://github.com/marketplace/actions/prml-verify)) for CI integration. MIT.
@@ -504,9 +504,9 @@ ln -sf "$(pwd)/hooks/commit-msg" .git/hooks/commit-msg
 ## Repository layout
 - `falsify.py` — single-file Python CLI, stdlib + pyyaml only.
-- `impl/js/falsify.js` — Node.js second reference implementation (12/12 v0.1 + 8/8 v0.2 = 20/20 vectors). Also published to npm as [`falsify-js`](https://www.npmjs.com/package/falsify-js).
-- `impl/go/falsify.go` — Go third reference implementation (20/20 vectors).
-- `impl/rust/` — Rust fourth reference implementation (20/20 vectors).
+- `impl/js/falsify.js` — Node.js second reference implementation (13/13 v0.1 + 8/8 v0.2 = 21/21 vectors). Also published to npm as [`falsify-js`](https://www.npmjs.com/package/falsify-js).
+- `impl/go/falsify.go` — Go third reference implementation (21/21 vectors).
+- `impl/rust/` — Rust fourth reference implementation (21/21 vectors).
 - `spec/PRML-v0.1.md` + `spec/test-vectors/v0.1/` (12) + `spec/test-vectors/v0.2/` (8) — spec + conformance suite.
 - `spec/analysis/` — positioning + canonicalization portability findings.
 - `spec/compliance/` — EU AI Act mapping + compliance landing copy.

{falsify-0.3.1 → falsify-0.3.2}/README.md RENAMED Viewed

@@ -20,14 +20,14 @@ TAMPERED  (exit 3)
 No install? Verify any manifest in-browser at [registry.falsify.dev](https://registry.falsify.dev). Byte-equivalent reference CLIs also ship for JS (`npm i -g falsify-js`), Go, and Rust.
-4 reference implementations (Python, JavaScript, Go, Rust) byte-equivalent on all 20 conformance vectors (12 v0.1 stable + 8 v0.2). PRML v0.2 frozen 2026-05-22. The same day, Lock #2 (a public hypothesis on the spec's own distribution, target ≥3 external contributors in 14 days) resolved at 0/3. The mechanism worked, the post-mortem is at [falsify.dev/notes/lock-2-postmortem](https://falsify.dev/notes/lock-2-postmortem/). Designed for ML eval rigor. Maps to EU AI Act Article 12 evidence as a side effect.
+4 reference implementations (Python, JavaScript, Go, Rust) byte-equivalent on all 21 conformance vectors (13 v0.1 stable + 8 v0.2). PRML v0.2 frozen 2026-05-22. The same day, Lock #2 (a public hypothesis on the spec's own distribution, target ≥3 external contributors in 14 days) resolved at 0/3. The mechanism worked, the post-mortem is at [falsify.dev/notes/lock-2-postmortem](https://falsify.dev/notes/lock-2-postmortem/). Designed for ML eval rigor. Maps to EU AI Act Article 12 evidence as a side effect.
 > **Pre-registration + CI for AI-agent claims.** Lock the claim and threshold with SHA-256 *before* running the experiment — or the result doesn't count.
 ![CI](https://github.com/studio-11-co/falsify/actions/workflows/falsify.yml/badge.svg)
 ![Multi-lang Conformance](https://github.com/studio-11-co/falsify/actions/workflows/multi-lang-conformance.yml/badge.svg)
 ![PyPI](https://img.shields.io/pypi/v/falsify?color=brightgreen&label=pypi)
-![coverage](https://img.shields.io/badge/tests-564%20passing-brightgreen)
+![coverage](https://img.shields.io/badge/tests-561%20passing-brightgreen)
 ![impls](https://img.shields.io/badge/reference%20impls-4%20(py%20%C2%B7%20js%20%C2%B7%20go%20%C2%B7%20rs)-brightgreen)
 ![honesty](https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/studio-11-co/falsify/main/.falsify/badge.json)
 ![python](https://img.shields.io/badge/python-3.11%2B-blue)
@@ -116,12 +116,12 @@ This repository is the home of **PRML v0.1** — Pre-Registered ML Manifest Spec
 **Audit & compliance crosswalks** (subcategory-by-subcategory maps from major AI governance frameworks to PRML fields, FULL/PARTIAL/NONE tagged):
-- **[EU AI Act Article 12](https://spec.falsify.dev/eu-ai-act/article-12/)** — code-level pattern for the 2 August 2026 high-risk applicability deadline
+- **[EU AI Act Article 12](https://spec.falsify.dev/eu-ai-act/article-12/)** — code-level pattern for the 2 December 2027 high-risk applicability deadline
 - **[Article 12 readiness diagnostic](https://spec.falsify.dev/article-12-readiness/)** — 10-question browser-only self-assessment
 - **[NIST AI RMF 1.0 crosswalk](https://spec.falsify.dev/nist-ai-rmf/)** — GOVERN / MAP / MEASURE / MANAGE subcategory map (incl. AI 600-1 GenAI Profile)
 - **[ISO/IEC 42001:2023 crosswalk](https://spec.falsify.dev/iso-42001/)** — AIMS clause-by-clause evidence map (Clauses 7-9 + Annex A controls)
-**Long-form working notes** (2026-05-23, written for compliance leads, AI governance officers, and notified body assessors preparing for the 2 August 2026 deadline; CC BY 4.0):
+**Long-form working notes** (2026-05-23, written for compliance leads, AI governance officers, and notified body assessors preparing for the 2 December 2027 deadline; CC BY 4.0):
 - **[EU AI Act readiness assessment](https://falsify.dev/eu-ai-act-readiness/)** — six binding articles, ten-question gap check, evidence shape per obligation
 - **[2 August 2026 deadline](https://falsify.dev/ai-act-deadline-august-2026/)** — three application dates, Article 99 penalty structure, ten-week plan
@@ -130,7 +130,7 @@ This repository is the home of **PRML v0.1** — Pre-Registered ML Manifest Spec
 - **[ISO/IEC 42001 readiness](https://falsify.dev/iso-42001-readiness/)** — seven clauses, EU AI Act Article 17 overlap, twelve-month certification path
 - **[Lock #2 post-mortem](https://falsify.dev/notes/lock-2-postmortem/)** — field report on running a falsifiable spec in public
-**Reference implementations** (four languages, 12 v0.1 + 8 v0.2 candidate vectors = 20 total; multi-lang CI runs all 20 byte-for-byte per push and daily at 04:00 UTC):
+**Reference implementations** (four languages, 13 v0.1 + 8 v0.2 candidate vectors = 21 total; multi-lang CI runs all 21 byte-for-byte per push and daily at 04:00 UTC):
 - **Python:** [`falsify.py`](falsify.py) — original reference, uses PyYAML
 - **Node.js:** [`impl/js/`](impl/js/) — second reference, ~400 LOC, hand-rolled, zero deps
@@ -141,7 +141,7 @@ Hosted spec at [spec.falsify.dev/v0.1](https://spec.falsify.dev/v0.1). Public re
 **Companion projects** (separate repos under `studio-11-co`, each MIT or CC0 licensed):
-- **[`falsify-cookbook`](https://github.com/studio-11-co/falsify-cookbook)** — field manual for the spec: 11 patterns + 4 anti-patterns, every one a single page with a runnable example, including [Pattern 11: PRML + Sigstore for execution integrity](https://github.com/studio-11-co/falsify-cookbook/blob/main/patterns/11-sigstore-execution.md) closing the §8.1 gap. CC0.
+- **[`falsify-cookbook`](https://github.com/studio-11-co/falsify-cookbook)** — field manual for the spec: 13 patterns + 4 anti-patterns, every one a single page with a runnable example, including [Pattern 11: PRML + Sigstore for execution integrity](https://github.com/studio-11-co/falsify-cookbook/blob/main/patterns/11-sigstore-execution.md) closing the §8.1 gap. CC0.
 - **[`falsify-integrity-index`](https://github.com/studio-11-co/falsify-integrity-index)** — public scorecard of how 25+ well-known ML eval claims meet the 9 PRML falsifiability criteria. Live at [falsify.dev/integrity](https://falsify.dev/integrity). CC0 data, MIT tooling.
 - **[`falsify-inspect`](https://github.com/studio-11-co/falsify-inspect)** — Inspect AI adapter: anchor an Inspect AI eval claim's threshold to a SHA-256 hash before the run, verify the post-run log against it. MIT.
 - **[`prml-verify-action`](https://github.com/studio-11-co/prml-verify-action)** — composite GitHub Action ([listed on Marketplace](https://github.com/marketplace/actions/prml-verify)) for CI integration. MIT.
@@ -472,9 +472,9 @@ ln -sf "$(pwd)/hooks/commit-msg" .git/hooks/commit-msg
 ## Repository layout
 - `falsify.py` — single-file Python CLI, stdlib + pyyaml only.
-- `impl/js/falsify.js` — Node.js second reference implementation (12/12 v0.1 + 8/8 v0.2 = 20/20 vectors). Also published to npm as [`falsify-js`](https://www.npmjs.com/package/falsify-js).
-- `impl/go/falsify.go` — Go third reference implementation (20/20 vectors).
-- `impl/rust/` — Rust fourth reference implementation (20/20 vectors).
+- `impl/js/falsify.js` — Node.js second reference implementation (13/13 v0.1 + 8/8 v0.2 = 21/21 vectors). Also published to npm as [`falsify-js`](https://www.npmjs.com/package/falsify-js).
+- `impl/go/falsify.go` — Go third reference implementation (21/21 vectors).
+- `impl/rust/` — Rust fourth reference implementation (21/21 vectors).
 - `spec/PRML-v0.1.md` + `spec/test-vectors/v0.1/` (12) + `spec/test-vectors/v0.2/` (8) — spec + conformance suite.
 - `spec/analysis/` — positioning + canonicalization portability findings.
 - `spec/compliance/` — EU AI Act mapping + compliance landing copy.

{falsify-0.3.1 → falsify-0.3.2/falsify.egg-info}/PKG-INFO RENAMED Viewed

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: falsify
-Version: 0.3.1
+Version: 0.3.2
 Summary: PRML reference CLI — pre-register an ML evaluation claim as a SHA-256 manifest; verify PASS/FAIL/TAMPERED.
 Author: Cüneyt Öztürk
 License: MIT
@@ -52,14 +52,14 @@ TAMPERED  (exit 3)
 No install? Verify any manifest in-browser at [registry.falsify.dev](https://registry.falsify.dev). Byte-equivalent reference CLIs also ship for JS (`npm i -g falsify-js`), Go, and Rust.
-4 reference implementations (Python, JavaScript, Go, Rust) byte-equivalent on all 20 conformance vectors (12 v0.1 stable + 8 v0.2). PRML v0.2 frozen 2026-05-22. The same day, Lock #2 (a public hypothesis on the spec's own distribution, target ≥3 external contributors in 14 days) resolved at 0/3. The mechanism worked, the post-mortem is at [falsify.dev/notes/lock-2-postmortem](https://falsify.dev/notes/lock-2-postmortem/). Designed for ML eval rigor. Maps to EU AI Act Article 12 evidence as a side effect.
+4 reference implementations (Python, JavaScript, Go, Rust) byte-equivalent on all 21 conformance vectors (13 v0.1 stable + 8 v0.2). PRML v0.2 frozen 2026-05-22. The same day, Lock #2 (a public hypothesis on the spec's own distribution, target ≥3 external contributors in 14 days) resolved at 0/3. The mechanism worked, the post-mortem is at [falsify.dev/notes/lock-2-postmortem](https://falsify.dev/notes/lock-2-postmortem/). Designed for ML eval rigor. Maps to EU AI Act Article 12 evidence as a side effect.
 > **Pre-registration + CI for AI-agent claims.** Lock the claim and threshold with SHA-256 *before* running the experiment — or the result doesn't count.
 ![CI](https://github.com/studio-11-co/falsify/actions/workflows/falsify.yml/badge.svg)
 ![Multi-lang Conformance](https://github.com/studio-11-co/falsify/actions/workflows/multi-lang-conformance.yml/badge.svg)
 ![PyPI](https://img.shields.io/pypi/v/falsify?color=brightgreen&label=pypi)
-![coverage](https://img.shields.io/badge/tests-564%20passing-brightgreen)
+![coverage](https://img.shields.io/badge/tests-561%20passing-brightgreen)
 ![impls](https://img.shields.io/badge/reference%20impls-4%20(py%20%C2%B7%20js%20%C2%B7%20go%20%C2%B7%20rs)-brightgreen)
 ![honesty](https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/studio-11-co/falsify/main/.falsify/badge.json)
 ![python](https://img.shields.io/badge/python-3.11%2B-blue)
@@ -148,12 +148,12 @@ This repository is the home of **PRML v0.1** — Pre-Registered ML Manifest Spec
 **Audit & compliance crosswalks** (subcategory-by-subcategory maps from major AI governance frameworks to PRML fields, FULL/PARTIAL/NONE tagged):
-- **[EU AI Act Article 12](https://spec.falsify.dev/eu-ai-act/article-12/)** — code-level pattern for the 2 August 2026 high-risk applicability deadline
+- **[EU AI Act Article 12](https://spec.falsify.dev/eu-ai-act/article-12/)** — code-level pattern for the 2 December 2027 high-risk applicability deadline
 - **[Article 12 readiness diagnostic](https://spec.falsify.dev/article-12-readiness/)** — 10-question browser-only self-assessment
 - **[NIST AI RMF 1.0 crosswalk](https://spec.falsify.dev/nist-ai-rmf/)** — GOVERN / MAP / MEASURE / MANAGE subcategory map (incl. AI 600-1 GenAI Profile)
 - **[ISO/IEC 42001:2023 crosswalk](https://spec.falsify.dev/iso-42001/)** — AIMS clause-by-clause evidence map (Clauses 7-9 + Annex A controls)
-**Long-form working notes** (2026-05-23, written for compliance leads, AI governance officers, and notified body assessors preparing for the 2 August 2026 deadline; CC BY 4.0):
+**Long-form working notes** (2026-05-23, written for compliance leads, AI governance officers, and notified body assessors preparing for the 2 December 2027 deadline; CC BY 4.0):
 - **[EU AI Act readiness assessment](https://falsify.dev/eu-ai-act-readiness/)** — six binding articles, ten-question gap check, evidence shape per obligation
 - **[2 August 2026 deadline](https://falsify.dev/ai-act-deadline-august-2026/)** — three application dates, Article 99 penalty structure, ten-week plan
@@ -162,7 +162,7 @@ This repository is the home of **PRML v0.1** — Pre-Registered ML Manifest Spec
 - **[ISO/IEC 42001 readiness](https://falsify.dev/iso-42001-readiness/)** — seven clauses, EU AI Act Article 17 overlap, twelve-month certification path
 - **[Lock #2 post-mortem](https://falsify.dev/notes/lock-2-postmortem/)** — field report on running a falsifiable spec in public
-**Reference implementations** (four languages, 12 v0.1 + 8 v0.2 candidate vectors = 20 total; multi-lang CI runs all 20 byte-for-byte per push and daily at 04:00 UTC):
+**Reference implementations** (four languages, 13 v0.1 + 8 v0.2 candidate vectors = 21 total; multi-lang CI runs all 21 byte-for-byte per push and daily at 04:00 UTC):
 - **Python:** [`falsify.py`](falsify.py) — original reference, uses PyYAML
 - **Node.js:** [`impl/js/`](impl/js/) — second reference, ~400 LOC, hand-rolled, zero deps
@@ -173,7 +173,7 @@ Hosted spec at [spec.falsify.dev/v0.1](https://spec.falsify.dev/v0.1). Public re
 **Companion projects** (separate repos under `studio-11-co`, each MIT or CC0 licensed):
-- **[`falsify-cookbook`](https://github.com/studio-11-co/falsify-cookbook)** — field manual for the spec: 11 patterns + 4 anti-patterns, every one a single page with a runnable example, including [Pattern 11: PRML + Sigstore for execution integrity](https://github.com/studio-11-co/falsify-cookbook/blob/main/patterns/11-sigstore-execution.md) closing the §8.1 gap. CC0.
+- **[`falsify-cookbook`](https://github.com/studio-11-co/falsify-cookbook)** — field manual for the spec: 13 patterns + 4 anti-patterns, every one a single page with a runnable example, including [Pattern 11: PRML + Sigstore for execution integrity](https://github.com/studio-11-co/falsify-cookbook/blob/main/patterns/11-sigstore-execution.md) closing the §8.1 gap. CC0.
 - **[`falsify-integrity-index`](https://github.com/studio-11-co/falsify-integrity-index)** — public scorecard of how 25+ well-known ML eval claims meet the 9 PRML falsifiability criteria. Live at [falsify.dev/integrity](https://falsify.dev/integrity). CC0 data, MIT tooling.
 - **[`falsify-inspect`](https://github.com/studio-11-co/falsify-inspect)** — Inspect AI adapter: anchor an Inspect AI eval claim's threshold to a SHA-256 hash before the run, verify the post-run log against it. MIT.
 - **[`prml-verify-action`](https://github.com/studio-11-co/prml-verify-action)** — composite GitHub Action ([listed on Marketplace](https://github.com/marketplace/actions/prml-verify)) for CI integration. MIT.
@@ -504,9 +504,9 @@ ln -sf "$(pwd)/hooks/commit-msg" .git/hooks/commit-msg
 ## Repository layout
 - `falsify.py` — single-file Python CLI, stdlib + pyyaml only.
-- `impl/js/falsify.js` — Node.js second reference implementation (12/12 v0.1 + 8/8 v0.2 = 20/20 vectors). Also published to npm as [`falsify-js`](https://www.npmjs.com/package/falsify-js).
-- `impl/go/falsify.go` — Go third reference implementation (20/20 vectors).
-- `impl/rust/` — Rust fourth reference implementation (20/20 vectors).
+- `impl/js/falsify.js` — Node.js second reference implementation (13/13 v0.1 + 8/8 v0.2 = 21/21 vectors). Also published to npm as [`falsify-js`](https://www.npmjs.com/package/falsify-js).
+- `impl/go/falsify.go` — Go third reference implementation (21/21 vectors).
+- `impl/rust/` — Rust fourth reference implementation (21/21 vectors).
 - `spec/PRML-v0.1.md` + `spec/test-vectors/v0.1/` (12) + `spec/test-vectors/v0.2/` (8) — spec + conformance suite.
 - `spec/analysis/` — positioning + canonicalization portability findings.
 - `spec/compliance/` — EU AI Act mapping + compliance landing copy.

{falsify-0.3.1 → falsify-0.3.2}/falsify.py RENAMED Viewed

@@ -25,7 +25,7 @@ from typing import Any, Callable
 import yaml
-__version__ = "0.3.1"
+__version__ = "0.3.2"
 EXIT_PASS = 0
 EXIT_FAIL = 10

{falsify-0.3.1 → falsify-0.3.2}/falsify_prml.py RENAMED Viewed

@@ -33,7 +33,7 @@ import os
 import re
 import sys
-__version__ = "0.3.1"
+__version__ = "0.3.2"
 EXIT_PASS = 0
 EXIT_BAD = 2

{falsify-0.3.1 → falsify-0.3.2}/pyproject.toml RENAMED Viewed

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 [project]
 name = "falsify"
-version = "0.3.1"
+version = "0.3.2"
 description = "PRML reference CLI — pre-register an ML evaluation claim as a SHA-256 manifest; verify PASS/FAIL/TAMPERED."
 readme = "README.md"
 license = { text = "MIT" }

{falsify-0.3.1 → falsify-0.3.2}/tests/test_prml_vectors.py RENAMED Viewed

@@ -5,7 +5,7 @@ manifest, the canonical UTF-8 bytes the canonicalizer MUST produce, and
 the SHA-256 hex digest of those bytes.
 If any of these fail, either:
-  (a) the canonicalizer in falsify._canonicalize has changed in a way
+  (a) the canonicalizer in falsify_prml.canonicalize has changed in a way
       that breaks the v0.1 contract — bump to v0.2 instead, OR
   (b) test vectors must be regenerated by running
       `python3 spec/test-vectors/v0.1/generate.py`.
@@ -25,7 +25,11 @@ REPO_ROOT = Path(__file__).resolve().parent.parent
 VECTORS_PATH = REPO_ROOT / "spec" / "test-vectors" / "v0.1" / "test-vectors.json"
 sys.path.insert(0, str(REPO_ROOT))
-import falsify  # noqa: E402
+# The PRML conformance suite verifies the PRML manifest canonicalizer — the
+# `falsify` command (falsify_prml), NOT the falsify-engine workflow canonicalizer
+# (falsify.py), which canonicalizes a different schema and does not apply the
+# PRML v0.1 threshold-float coercion.
+import falsify_prml  # noqa: E402
 def _load_vectors():
@@ -46,8 +50,10 @@ class PRMLVectorTests(unittest.TestCase):
             raise unittest.SkipTest(f"test vectors not generated yet: {VECTORS_PATH}")
     def test_vector_count(self):
-        """v0.1 ships with exactly 12 test vectors. Adding more is a v0.2 change."""
-        self.assertEqual(len(VECTORS), 12, f"Expected 12 vectors, got {len(VECTORS)}")
+        """v0.1 ships 13 test vectors (TV-001..TV-013). TV-013 locks the
+        integer-threshold->float coercion; it is a non-breaking, clarifying
+        addition that leaves every prior vector's hash unchanged."""
+        self.assertEqual(len(VECTORS), 13, f"Expected 13 vectors, got {len(VECTORS)}")
     def test_key_order_invariance(self):
         """TV-001 and TV-002 differ only in input key insertion order. Hashes MUST match."""
@@ -83,7 +89,7 @@ class PRMLVectorTests(unittest.TestCase):
 def _make_canonical_test(vector):
     def test(self):
-        produced = falsify._canonicalize(vector["input"])
+        produced = falsify_prml.canonicalize(vector["input"])
         self.assertEqual(
             produced,
             vector["canonical"],
@@ -99,7 +105,7 @@ def _make_canonical_test(vector):
 def _make_hash_test(vector):
     def test(self):
-        canonical = falsify._canonicalize(vector["input"])
+        canonical = falsify_prml.canonicalize(vector["input"])
         produced_hash = hashlib.sha256(canonical.encode("utf-8")).hexdigest()
         self.assertEqual(
             produced_hash,

{falsify-0.3.1 → falsify-0.3.2}/tests/test_version.py RENAMED Viewed

@@ -35,14 +35,14 @@ class VersionTests(unittest.TestCase):
     def test_version_subcommand_prints_version(self) -> None:
         result = _run(["version"], cwd=self.cwd)
         self.assertEqual(result.returncode, 0, msg=result.stderr)
-        self.assertIn("0.3.1", result.stdout)
+        self.assertIn("0.3.2", result.stdout)
         self.assertIn("falsify", result.stdout)
     def test_version_flag_prints_version(self) -> None:
         result = _run(["--version"], cwd=self.cwd)
         self.assertEqual(result.returncode, 0, msg=result.stderr)
         # argparse's `action='version'` writes to stdout on Python 3.11+.
-        self.assertIn("0.3.1", result.stdout)
+        self.assertIn("0.3.2", result.stdout)
     def test_version_subcommand_json_mode(self) -> None:
         result = _run(["version", "--json"], cwd=self.cwd)