kimi/Timmy-time-dashboard
Archived
1
0
Fork 0
forked from Rockachopa/Timmy-time-dashboard
Code Pull Requests Activity

[claude] Add content moderation pipeline (Llama Guard + game-context prompts) (#1056) (#1059)

Browse Source
This commit is contained in:
Claude (Opus 4.6)
2026-03-23 02:14:42 +00:00
parent 092c982341
commit 1697e55cdb
6 changed files with 1013 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
src/infrastructure/guards/__init__.py Normal file
View File

@@ -0,0 +1,7 @@
"""Content moderation pipeline for AI narrator output.
Three-layer defense:
1. Game-context system prompts (vocabulary whitelists, theme framing)
2. Real-time output filter via Llama Guard (or fallback regex)
3. Per-game moderation profiles with configurable thresholds
"""