michelligabriele ae454fd700 fix(proxy): OpenAI Moderation post-call guardrail response not captured for logging ... Two independent bugs prevented post-call OpenAI Moderation guardrail results from reaching downstream logging callbacks (Langfuse, Datadog). Bug 1: process_output_response() created a throwaway request_data dict, so guardrail info written by @log_guardrail_information was discarded. Fixed by threading the real request_data from the unified guardrail dispatcher through all 13 BaseTranslation handlers, with litellm_metadata injection preserved for third-party guardrails (Zscaler, Prompt Security). Also extended to process_output_streaming_response for consistency. Bug 2: The @log_guardrail_information decorator collapsed the full moderation API response (categories, scores, flagged status) to "allow". Fixed by overriding _process_response/_process_error on OpenAIModerationGuardrail to stash and log the full response, following the established Model Armor pattern.		2026-03-23 15:39:22 +01:00
..
mcp_server	fix(proxy): OpenAI Moderation post-call guardrail response not captured for logging	2026-03-23 15:39:22 +01:00
out	chore: update Next.js build artifacts (2026-03-22 05:53 UTC, node v22.16.0)	2026-03-21 22:53:50 -07:00
post_call_rules.py	refactor: add black formatting	2023-12-25 14:11:20 +05:30