v1.0.1

Changed

• SKILL.md: removed the nested metadata.openclaw block — the agentskills
  spec (enforced by awesome-copilot's vally linter) requires metadata values
  to be strings; tool requirements (gh, git, jq, python3) are now
  declared in the Preflight section instead
• SKILL.md Templates list: qualified README.md / README.zh.md as
  templates/README.* so file-reference linters resolve them correctly

v1.0.0

Added

• SKILL.md: explicit template-placeholder substitution step — OWNER/REPO,
  OWNER_USERNAME, PROJECT_NAME, YYYY-MM-DD are replaced with the
  target repo's real values when templates are copied (badges, discussions,
  security advisory, clone URLs no longer ship broken)
• SKILL.md metadata: requires now lists jq + python3 alongside
  gh / git (they were already needed by Module A)
• SKILL.md Module D: documents $default-branch triggers and lockfile-aware
  dependency install for Node projects
• SKILL.md Module B: issue forms now also follow the user's chosen language
  (Preflight step 5)

Changed

• CI templates: push.branches: [main] → [$default-branch] — workflows
  now trigger on any default branch name (Node / Python / Go / Rust)
• ci-node.yml: dependency install detects the lockfile (npm ci /
  pnpm install --frozen-lockfile / yarn install --frozen-lockfile /
  plain npm install fallback)
• ci-go.yml: matrix updated to Go 1.25 / 1.26 (1.22 / 1.23 were EOL)
• templates/labels.json: P3 color 0e8a16 → 1a7f37 so it no longer
  duplicates status: ready to merge's color
• templates/CONTRIBUTING.md: PR target wording no longer hardcodes main
• VISION.md: current state reflects v1.0
• Repo labels dogfooded to the v1 baseline: P0–P3 and impact:* added

Removed

• Legacy priority: critical/high/medium/low labels from this repo
  (superseded by P0–P3)

v0.2.8

Fixed

• CI templates: removed the npm-only --if-present flag from the Python
  (ruff check . / pytest) and Rust (cargo fmt --check / cargo clippy)
  workflows — those tools reject the flag, which broke CI for non-npm projects
• templates/CONTRIBUTING.md: issue-label reference now matches the actual
  taxonomy (bug / enhancement / P0–P3) instead of a non-existent
  type:* / priority:* prefix
• templates/labels.json: impact: data-loss and impact: availability no
  longer reuse impact: security's color — fixes the adjacent-duplicate that
  the skill's own color rules forbid

Changed

• SKILL.md / templates/LABELS.md: dropped stray Chinese glosses (段位 / 分工小组)
  from the English docs

v0.2.7

Changed

• Surface-only positioning: SKILL.md description + intro, README (en + zh)
  tagline and Safety-by-design now state clearly that the skill polishes a
  repo's surface (labels, templates, CI configs, rulesets, docs) and never
  touches code logic; GitHub repo description updated with the same wording
• Repo moved from programmingHLS to programmingWTF (personal account, per
  Guiyu's call — consistent with ClawHub publisher identity); all path
  references updated (README badges/install/usage, issue-form config,
  SECURITY.md, AGENTS.md); old URL redirects automatically

v0.2.6

Changed

• Emoji policy relaxed: rating tiers (rating:* / issue-rating:*) are now
  the ONLY mandatory-emoji labels (clear low→high gradient); all other
  labels leave emoji to the agent's judgment
• Consistency rule added: within one dimension, either ALL labels carry an
  emoji or NONE do — never a mixed half-emoji dimension
• templates/labels.json baseline now ships plain names (no emoji) —
  bug, P0–P3, status: *, dependencies, security…; impact labels
  were already emoji-free, so every dimension is now consistent
• SKILL.md / LABELS.md / README (en + zh) wording updated to the new policy

v0.2.5

Changed

• description updated to reflect the fuller taxonomy: P0–P3 priority, impact
  severity, auto-close rules
• SKILL.md Module A: bot-label language exception — r:* / clawsweeper:* /
  triage:* / close:* names stay in the bot's language (automation matches
  literal names); only their description uses the user's language

v0.2.4

Added

• Preflight step 6: ask whether the repo has automation bots / AI writers
  (dependabot, Stale, ClawSweeper, AI coding agent…) — the answer decides
  whether the Governance / auto-close-rules labels are added at all, with the
  reasoning spelled out
• Module A Governance dimension now gated by Preflight step 6

v0.2.3

Added

• SKILL.md Module A: Governance / auto-close rules dimension — r:* and
  bot-state labels (clawsweeper:*, triage:*, close:*) are signal
  labels, not categories, only meaningful when a governance bot actually
  runs on the repo; skip them otherwise

v0.2.2

Changed

• Priority labels switched to P0–P3 (OpenClaw convention; P0 = emergency)
• Added impact: severity dimension — security / data-loss / availability /
  auth-provider / session-state / message-loss / ux-blocker / ux-friction
• SKILL.md Module A rewritten into a full dimension menu mirroring OpenClaw's
  official taxonomy — every dimension is opt-in ("add only if the repo needs
  it"), Type is the only mandatory one
• Label language rule: names + descriptions follow the user's chosen language
  (Preflight step 5)

v0.2.1

Changed

• Label taxonomy now uses emoji prefixes across every category (type,
  priority, status, team, rating) — not just rating tiers — so the whole set
  is scannable at a glance
• SKILL.md: emoji-prefix rule promoted from rating-only to the full taxonomy;
  Module A tables and examples updated
• SKILL.md description + README tagline rewritten for discoverability — adds
  professionalize, tidy up, polish, emoji label taxonomy trigger words

Added

• templates/labels.json: emoji-prefixed baseline labels (🐛 ✨ 📚 ❓ 🙋 🌱
  🔴🟠🟡🟢 🧱🚧✅🎉🚫 📦 🔒)