update for mcp

2026-08-06 11:08:46 +08:00
parent 31bbf80aeb
commit b2feaeddb4
40 changed files with 1986 additions and 202 deletions
@@ -17,6 +17,11 @@ beautifulsoup4>=4.12.0
 lxml>=5.0.0
 tiktoken>=0.5.0
 tenacity>=8.2.0
+# Regulatory signal crawling (backend/app/infrastructure/perception/) — main-content
+# extraction from crawled regulation detail pages and character-level diff for change
+# detection. Import name for diff-match-patch is diff_match_patch (underscored).
+trafilatura>=2.0.0
+diff-match-patch>=20241021

 # ── Auth ──────────────────────────────────────────────────────────────────────
 python-jose[cryptography]>=3.3.0