Skip to content

CHANGELOG.md

Latest commit

 

History

History
58 lines (48 loc) · 2.95 KB

CHANGELOG.md

File metadata and controls

58 lines (48 loc) · 2.95 KB
Raw

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog.

[0.3.0] - 2026-03-24

Changed

  • Core framework replaced: "Three Represents" → "Serve the People" (为人民服务) — Mao Zedong's 1944 speech with five principles (全心全意, 实事求是, 极端负责, 极端热忱, 不怕批评)
  • Version bumped to 0.3.0
  • README/README_zh-cn: "Why Three Represents?" → "Why Serve the People?" with new comparison table
  • SKILL.md: All check dimensions rewritten to align with the five principles
  • Philosophy docs completely rewritten with Mao's original text quotes
  • Example reports updated with new dimension names and Chinese annotations
  • CONTRIBUTING.md: references updated from Three Represents to Serve the People
  • Comparison table: "Three Represents" → "Serve the People framework"
  • Emotional Temperature dimension renamed to "Emotional Intelligence" for clarity

[0.2.0] - 2026-03-24

Added

  • Option B: Universal LLM usage — soul-doctor now works without OpenClaw, with any LLM
  • Sample workspace in examples/sample-workspace/ for instant trial
  • Step 0: Minimum Viable Workspace — soul-doctor now checks for SOUL.md before running
  • CI/CD — GitHub Actions for markdown linting, link checking, and content consistency
  • CODE_OF_CONDUCT.md — Contributor Covenant
  • CHANGELOG.md — this file
  • Dynamic GitHub badges (stars, last commit, issues)
  • memory/ filename convention (YYYY-MM-DD.md)
  • Chinese character count target alongside English word count
  • CLAUDE.md added to the 11-file completeness checklist

Fixed

  • Off-by-one bug: file completeness listed 10 items but report template said "X/11" — now correctly lists 11 items
  • Trigger phrase mismatch: "check my soul" was in README but missing from SKILL.md YAML frontmatter
  • Grading logic conflict: C and D grades both included "Values Consistency ❌" — now clearly differentiated
  • Comparison table fairness: Generic AI audit column now shows "Partial" for values check instead of all dashes

Changed

  • Version bumped to 0.2.0
  • Quick Start split into Option A (OpenClaw) and Option B (Any LLM)
  • Step 1 now clearly separates "read for content" (7 files) from "check existence only" (4 additional items)
  • Token Efficiency now specifies <3000 words (EN) / <5000 chars (CN) for SOUL.md
  • README badges changed from all-static to dynamic GitHub badges
  • Philosophy docs: "Too Western" replaced with more constructive framing

[0.1.0] - 2026-03-24

Added

  • Initial release
  • SKILL.md — Core skill definition with 4-step check flow
  • 7 evaluation dimensions in 3 tiers (Three Represents, Character, Technical)
  • Grading rubric (A+ through D)
  • Bilingual documentation (English + Chinese)
  • Example reports (A+ and D grades)
  • CONTRIBUTING.md with contribution guidelines
  • GitHub issue templates (bug report, feature request, report sharing)