- compare: add NO COMPARATIVE RANKING and NO FABRICATED DIALOGUE
iron laws with rationalization defense table
- create-master: add NO FABRICATED SOURCES iron law for the
generation pipeline
- Add 2 boundary tests for compare (sectarian ranking, fabricated
dialogue)
- Fix validate-fidelity.py to support compare's assertion fields
and no_fabricated_dialogue boundary type
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Inspired by obra/superpowers project patterns, add three-layer defense
to prevent AI hallucination and rule violation:
- Iron Laws: NO DOCTRINAL CLAIM WITHOUT CBETA CITATION,
NO PERSONA BEFORE CONTEXT, NO SECTARIAN JUDGMENT
- Rationalization Defense Table: 6 common AI excuses with rebuttals
- Red Flag List: 5 observable violation signals
Also add 40 pressure test cases (5 per master) covering:
- Sectarian judgment boundary
- No prophecy boundary
- Neutral first turn boundary
- Citation bypass pressure
- Master-specific challenges (meta, hostile, misunderstanding, etc.)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Previously LLM might run ONE combined semantic query, causing the non-dominant
master to get fallback search URLs (fojin.app/search?q=...) instead of precise
text_id links.
Now the skill:
- Explicitly requires N separate queries for N masters (emphasized with ⚠️)
- Provides a query-rewriting table: each master's query uses its tradition's
terminology (唯识/中观/禅宗/天台/华严/净土 terms)
- Example: user asks '因缘果' → xuanzang queries '因缘果 六因 四缘 五果 种子',
kumarajiva queries '因缘所生法 空 中道 无自性 四句'
This ensures both masters get precise text_id citations, not degraded search URLs.
- New --brief flag for semantic/search subcommands outputs one-line-per-result
(title + link + 80-char snippet) instead of full 500-char excerpts
- Reduces output from ~60 lines to ~7 lines per query (3 results)
- /compare-masters now uses --brief by default (multiple queries no longer
flood the terminal with long excerpts)
- Full output (default) preserved for single-master /ask and debugging
- Add Layer 0 hard rule to all 8 masters' voice.md: first turn must use
neutral address (您/汝/你/问者), forbidden terms include 居士/行者/学人/
善男子/善女人/出家人/师父/大众/道友/善信/道友
- From turn 2+: masters adapt to user's self-disclosed or question-inferred
identity, restoring each master's historical address style
- Layer 2 开场方式/称呼方式 reorganized into 首轮中立 / 身份已知后 tiers
- Update voice_builder.md and voice_analyzer.md templates so future
/create-master runs inherit this rule
- Add tools/sync_skill_from_voice.py to keep SKILL.md PART B in sync
- Add 48 regression tests in test_voice_rules.py (all 79 tests pass)
Each teacher's meta.json now includes search_scope with tradition-specific
CBETA/SuttaCentral IDs, dictionary sources, and keywords for focused RAG
retrieval. New tools/cross_reference.py enables lineage queries and
cross-tradition concept comparison via FoJin KG.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
xianren