Files
CASAN/packages/casan-harness/scripts/bash/unicode-normalize.py
T
thanhnvandClaude Opus 4.8 36a4812ef3 refactor(structure): promote app to repo root + remove redundant workspace cruft
Standard production layout: the OKR app (was nested under AINative_OKR_CASAN5/) is now
the repository root. No more wrapper directory.

- Promote AINative_OKR_CASAN5/* -> repo root (backend/ frontend/ packages/ apps/
  .specify/ docs/ infra/ nginx/ scripts/ + configs). Merge tool dirs: .gitea (kept the
  active deploy ci.yml, added harness-ci.yml + runbooks), .claude (agents/commands +
  launch.json), .github moved up.
- Remove redundant: 00_SUBMISSION_PACKAGE, scattered root notes (FPT_CASAN_Full.md,
  tu-tuong-casan.md, casan-tu-sinh..., casan_harness_assessment.md, source-review...,
  README_CASAN5_REFINED.md), casan-next-plans/ and optimize-docs/ (competition/planning
  artifacts — roadmap + design history preserved in git log / commit messages).
- Update all references to the old layout:
  - .gitea/workflows/{ci,harness-ci}.yml, .github/workflows/{ci,deploy}.yml:
    working-directory .; drop AINative_OKR_CASAN5/ prefix; .specify/{tests,scripts}
    -> packages/casan-harness/... (.specify/logs state kept)
  - .claude/launch.json, .gitea/*-runbook.md: path prefixes
  - CLAUDE.md, README.md: docs/input -> apps/okr/domain/input
  - policy-bundle.yaml: 8 policy paths -> packages/casan-harness/...; manifest re-signed
- secrets-scan.sh: fixture excludes -> new package/domain paths.

Full gate from the new root: PASS=64 FAIL=0 SKIP=3.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-08 13:26:36 +09:00

63 lines
2.6 KiB
Python
Executable File
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
#!/usr/bin/env python3
"""CASAN H4 — Unicode confusable / obfuscation normalizer (Track A, V3).
Reads text on stdin, writes a normalized variant on stdout used ONLY for
injection/jailbreak phrase matching (never for PII/secret redaction, so we
never widen what counts as a secret).
Deterministic transforms, in order:
1. NFKC compatibility normalization — folds fullwidth (ignore) and
other compatibility forms to their ASCII equivalents.
2. Strip zero-width / BOM / soft-hyphen formatting characters that split a
word so a blocklist never sees it (i·g·n·o·r·e).
3. Fold a fixed table of common Cyrillic/Greek homoglyphs to their Latin
lookalikes (іgnоrе -> ignore). NFKC does NOT do this — confusables are a
separate Unicode concern — so the table is explicit and auditable.
The table is intentionally small and covers the lookalikes actually used in
prompt-injection homoglyph attacks; extend it as new vectors appear.
"""
import sys
import unicodedata
# Zero-width, BOM, and invisible formatting code points.
ZERO_WIDTH = {
0x200B, # zero-width space
0x200C, # zero-width non-joiner
0x200D, # zero-width joiner
0x2060, # word joiner
0xFEFF, # BOM / zero-width no-break space
0x00AD, # soft hyphen
0x180E, # Mongolian vowel separator
0x2061, 0x2062, 0x2063, 0x2064, # invisible math operators
}
# Common Cyrillic / Greek homoglyphs -> Latin lookalike. Lowercase and
# uppercase both listed because matching is case-insensitive downstream but the
# fold must run before case handling to be safe.
CONFUSABLES = {
# Cyrillic lowercase
"а": "a", "е": "e", "о": "o", "р": "p", "с": "c", "у": "y", "х": "x",
"ѕ": "s", "і": "i", "ј": "j", "ԁ": "d", "һ": "h", "ӏ": "l", "п": "n",
"г": "r", "т": "t", "к": "k", "м": "m", "в": "b",
# Cyrillic uppercase
"А": "A", "В": "B", "Е": "E", "К": "K", "М": "M", "Н": "H", "О": "O",
"Р": "P", "С": "C", "Т": "T", "У": "Y", "Х": "X", "Ѕ": "S", "І": "I",
"Ј": "J",
# Greek
"α": "a", "ε": "e", "ο": "o", "ρ": "p", "τ": "t", "ν": "v", "κ": "k",
"Α": "A", "Β": "B", "Ε": "E", "Ζ": "Z", "Η": "H", "Ι": "I", "Κ": "K",
"Μ": "M", "Ν": "N", "Ο": "O", "Ρ": "P", "Τ": "T", "Υ": "Y", "Χ": "X",
}
def normalize(text: str) -> str:
text = unicodedata.normalize("NFKC", text)
text = "".join(ch for ch in text if ord(ch) not in ZERO_WIDTH)
text = "".join(CONFUSABLES.get(ch, ch) for ch in text)
return text
if __name__ == "__main__":
sys.stdout.write(normalize(sys.stdin.read()))