Slash command discoverability cleanup. Claude Code users typically have
50+ skills installed; bare-word commands like /atisha and /zhiyi got
scattered across the /-completion list. v0.6 prefixes all 14 master
slash commands with master- so they cluster under /m<tab> and clearly
signal "this is a Master-skill master skill".
Scope of rename
---------------
* Slash commands: /zhiyi → /master-zhiyi, /huineng → /master-huineng,
... all 14 affected.
* Directory layout: prebuilt/<slug>/ → prebuilt/master-<slug>/ for all
14 masters (git mv preserves history).
* Frontmatter: each prebuilt/master-<slug>/SKILL.md updates `name:` to
master-<slug>.
* compare-masters and create-master meta-skills are intentionally
unchanged — they're already prefixed by their nature, and
/master-compare-masters would be doublespeak.
Decoupling: fojin.app/chat is NOT affected
------------------------------------------
The fojin web frontend's master dropdown uses bare slug IDs (atisha,
huineng, ...) and is already grouped under "法师模式" in its UI. Backend
master_profiles.py keeps `id="atisha"` etc. unchanged. No fojin-side
migration required. The two surfaces (Claude Code slash + fojin
dropdown) are now formally decoupled by design, not coincidence.
Compatibility
-------------
* NPX installer accepts both forms: `npx master-skill install zhiyi`
(short) and `install master-zhiyi` (full) both resolve to the same
prebuilt/master-zhiyi/ source. Install destination is always
~/.claude/skills/master-<slug>/. Backward-compatible uninstall
handles legacy non-prefixed installs (~/.claude/skills/zhiyi/).
* The cli.mjs already used `master-${name}` for install destinations
(since v0.3 NPX installer was added), so existing v0.4/v0.5 NPX
users were already getting the prefix in skills/ — only the source
prebuilt/ layout and slash commands change in v0.6.
Files updated
-------------
* 14 directories renamed (28 files moved, 0 content changes).
* 14 SKILL.md frontmatter `name:` fields.
* prebuilt/compare/SKILL.md: 43 slug references updated to prefixed form.
* bin/cli.mjs: resolveMasterDir helper accepts both short and full;
cmdInstall and cmdUninstall handle legacy paths.
* .github/workflows/validate-and-test.yml: fidelity-smoke MASTERS
rotation array updated to all 14 prefixed names (was 8 hardcoded
汉传 only — now properly rotates across the full set).
* scripts/{validate,cite,query,test-fidelity}.py: --master arg help
text examples.
* README.md + README_EN.md: situational guidance table, install
snippets, master cards. New v0.6 release banner.
* SKILL.md (project-level) preset list with new slash names.
* ETHICS.md Tier table slug references (4).
* All plugin manifests bumped 0.5.0 → 0.6.0 with description noting
the /master-<slug> invocation pattern.
* CHANGELOG.md: [0.6.0] section with breaking-change notice and
migration commands for existing NPX users.
Validation
----------
* python scripts/validate.py --strict ✓ 15 masters pass
* python scripts/validate-fidelity.py ✓ all valid
* python scripts/test-fidelity.py --all --dry-run ✓
* pytest tests/ ✓ 31 passed, 6 skipped
* node bin/cli.mjs list ✓ shows all 14 with
master- prefix
* node bin/cli.mjs install zhiyi ✓ resolves to
prebuilt/master-zhiyi/
* node bin/cli.mjs install master-zhiyi ✓ resolves to same
Migration for existing v0.4/v0.5 users
--------------------------------------
npx master-skill@0.5 uninstall zhiyi huineng xuanzang ...
# OR: rm -rf ~/.claude/skills/{zhiyi,huineng,...}
npx master-skill@latest install --all
Then start a new Claude Code session; the new slash commands are
/master-zhiyi, /master-huineng, etc.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Three governance-tier additions to convert Master-skill from a demo into a
defensible long-lived project:
**ETHICS.md** — mandatory governance document:
- AI transparency: outputs are AI-synthesized, not masters' own words
- Copyright tiers A (public domain, current 8), B (in-copyright, needs
license), C (never admit: living masters, Buddhas/bodhisattvas, apocryphal
figures), D (case-by-case)
- Religious boundary: AI must refuse precept transmission, awakening
certification, karmic diagnosis, spirit-medium framing, etc.
- Dual-track content license: code MIT, master content CC BY-NC-SA 4.0,
prompts CC BY 4.0
- Takedown + appeal channel with 48h / 7d SLAs
**CHANGELOG.md** — Keep a Changelog format:
- [Unreleased] captures the current governance + community + npm work
- [0.3.0] retroactively documents the architectural rebuild (provenance,
fidelity, NPX, multi-platform, HARD-GATE, two-stage review)
- [0.2.0], [0.1.0] historical sections
**CI fidelity smoke** — make HARD-GATE a real gate, not just documentation:
- New `fidelity-smoke` job runs one basic-difficulty fixture against one
master per PR; picks the master touched by the diff, else rotates by
day-of-year for uniform coverage
- Cost cap ≈ $0.05/PR (~10k-token system prompt × 1 request × Sonnet 4.6
pricing). Forks without ANTHROPIC_API_KEY get an advisory pass so
external PRs can still land
- `scripts/test-fidelity.py` gains `--max-tests N` flag; when capping, it
sorts by difficulty (basic → intermediate → advanced) so smoke runs hit
the reliable floor, not stress cases
- Old `fidelity` job renamed to `fidelity-full` (still workflow_dispatch)
README 声明 section now links to ETHICS.md so every reader sees the AI
disclosure and boundary rules before copying master content.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Update test-fidelity.py to support must_not_contain and
must_not_contain_first_turn fields for boundary/pressure tests
- Add validate-fidelity.py for structural validation of all
fidelity.jsonl files (no API needed)
- Add GitHub Actions workflow: runs validate + dry-run on every
push/PR touching prebuilt/scripts/prompts/tools
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
xianren