v1.12

What's Changed

• feat(metrics): Lexical Density (content vs. function token share) by @aspala in #68
• feat(metrics): Conditional Entropy & Markov-Perplexity by @aspala in #71
• feat(metrics): MTLD (Measure of Textual Lexical Diversity) by @aspala in #72
• fix(comments): sticky sentinel on agent-actions PR comment part by @aspala in #73

Full Changelog: v1.11...v1.12

v1.11

What's Changed

• perf(block-impact): byte cap + node-level parallelism (42.5s→15.4s) by @aspala in #67

Full Changelog: v1.10...v1.11

v1.10

What's Changed

• chore(deps-dev): bump credo from 1.7.18 to 1.7.19 in the minor-and-patch group by @dependabot[bot] in #64
• fix: JS/Phoenix false-positives in health-report (no_dead_code, boolean ?-suffix) by @aspala in #66

Full Changelog: v1...v1.10

v1.9

What's Changed

• perf(health-report): skip codebase metrics in base snapshot by @aspala in #63

Full Changelog: v1.8...v1.9

v1.8

What's Changed

• perf(block-impact): byte-exact block slice enables subtractive LOO by @aspala in #62

Full Changelog: v1.7...v1.8

v1.7

What's Changed

• perf(health-report): skip leave-one-out in base snapshot by @aspala in #61

Full Changelog: v1...v1.7

v1

What's Changed

• feat: add self-hosted health-report workflow for PRs by @aspala in #1

New Contributors

• @aspala made their first contribution in #1

Full Changelog: https://github.com/num42/codeqa-action/commits/v1

What's Changed

• fix: add Erlang/OTP setup step — escript requires runtime by @aspala in #2

Full Changelog: v0.1.0...v1

What's Changed

• fix: add Erlang/OTP setup step — escript requires runtime by @aspala in #2

Full Changelog: v0.1.0...v1

What's Changed

• fix(compare): gracefully handle compare when no source files changed by @aspala in #4

Full Changelog: v1.1.0...v1

What's Changed

• fix(compare): gracefully handle compare when no source files changed by @aspala in #4

Full Changelog: v1.1.0...v1

What's Changed

• chore: formatting, moduledocs, and gitignore cleanup by @aspala in #5
• ci: add test workflow for PRs by @aspala in #7
• feat(health-report): add github format with mermaid charts and progress bars by @aspala in #3

Full Changelog: v1.1.1...v1

What's Changed

• feat: worst offenders table extensions by @aspala in #8
• feat(ci): add compare workflow for PR code quality diff by @aspala in #9

Full Changelog: v1.2...v1

What's Changed

• feat(health-report): split into --view metrics|actions|both + skip unneeded work by @aspala in #59
• feat(metrics): Rényi entropy spectrum & Hill numbers (#46) by @aspala in #57
• fix(action): scope health-report to the PR diff via base-ref by @aspala in #60

Full Changelog: v1.5...v1

What's Changed

• perf(health-report): skip leave-one-out in base snapshot by @aspala in #61

Full Changelog: v1.6...v1

What's Changed

• perf(block-impact): byte-exact block slice enables subtractive LOO by @aspala in #62

Full Changelog: v1.7...v1

What's Changed

• perf(health-report): skip codebase metrics in base snapshot by @aspala in #63

Full Changelog: v1.8...v1

v1.6

What's Changed

Health-report split into --view metrics | actions | both (#59)

The report's two responsibilities are now separated behind a --view flag (default both):

• metrics — numberified metric scales only (categories, cosine behavior scores, top issues)
• actions — agent-consumable refactoring prompt with YAML frontmatter; critical/high blocks rendered as numbered tasks with concrete, templated fix hints
• both — metrics followed by the agent-actions prompt (replaces the old block table)

Each view skips work it doesn't need: metrics never runs the block-impact leave-one-out (the heaviest phase), and actions skips the base snapshot (a full second analysis run). Every behavior now carries a concrete _fix_hint (templated with block context) instead of the old "Review this code block" fallback.

Rényi entropy spectrum & Hill numbers (#57, closes #46)

New Metrics.File.RenyiEntropy adds a Rényi-entropy spectrum (orders 0/1/2/∞), Hill numbers (effective vocabulary size), and a concentration slope — a more robust characterization of token-distribution shape than a single Shannon scalar. Woven into all behavior vectors, complementing Lexical Concentration.

Full Changelog: v1.5...v1.6

What's Changed

• feat(health-report): split into --view metrics|actions|both + skip unneeded work by @aspala in #59
• feat(metrics): Rényi entropy spectrum & Hill numbers (#46) by @aspala in #57

Full Changelog: v1.5...v1.6

What's Changed

• feat(health-report): split into --view metrics|actions|both + skip unneeded work by @aspala in #59
• feat(metrics): Rényi entropy spectrum & Hill numbers (#46) by @aspala in #57
• fix(action): scope health-report to the PR diff via base-ref by @aspala in #60

Full Changelog: v1.5...v1.6

v1.5

What's Changed

• fix(metrics) + perf(health-report): correct separator LOO, scope block-impact to changed files by @aspala in #58

Full Changelog: v1.4...v1.5

v1.4

What's Changed

• feat(tooling): add dialyxir and fix all dialyzer warnings by @aspala in #12
• refactor(metrics): improve all metrics with bug fixes, new fields, and typespecs by @aspala in #13
• fix(compare): github format always showed 'no changes — all metrics stable' by @aspala in #15
• feat(cli): auto-load ignore_paths from .codeqa.yml by @aspala in #16
• refactor(cli): split monolithic cli.ex into focused command modules by @aspala in #14
• docs(readme): rewrite with full CLI reference, metrics, and grading by @aspala in #21
• feat(tooling): add mix precommit alias and devenv git hook by @aspala in #19
• feat(github): add structured issue templates with bot mirror support by @aspala in #23
• ci(dialyzer): skip when no Elixir files changed by @aspala in #24
• feat(compare): default to all-files for complete PR comparison by @aspala in #25
• chore(samples): add behavior sample fixtures across languages by @aspala in #26
• feat(engine): near-duplicate-blocks engine and layered architecture by @aspala in #28
• feat(scalar-tuner): add Vite/React UI for behavior threshold tuning by @aspala in #27
• chore: introduce num42_refactors + apply safe AST refactors by @aspala in #30
• fix(health-report): suppress false-positive per-block findings by @aspala in #35
• fix(health-report): suppress single-block phantoms + de-overfit samples by @aspala in #36
• fix(health-report): add _excludes_languages + de-overfit FP behaviors by @aspala in #37
• chore(deps): bump the minor-and-patch group with 5 updates by @dependabot[bot] in #34
• chore(deps): bump the all-actions group with 4 updates by @dependabot[bot] in #33
• chore(deps): bump vite to 8.0.16 in scalar_tuner by @aspala in #38
• fix(health-report): exclude doc blocks from loop_var + declared_close_to_use by @aspala in #39
• chore(samples): remove 207 non-Elixir behavior sample dirs by @aspala in #40
• fix(health-report): kill no_debug_print HIGH false positive on data modules by @aspala in #41
• feat(metrics): Lexical Concentration (Yule's K / Simpson's D) by @aspala in #56

New Contributors

• @dependabot[bot] made their first contribution in #34

Full Changelog: v1...v1.4