mirror of
https://github.com/BerriAI/litellm.git
synced 2026-09-14 23:21:35 +00:00
The per-chunk streaming hook skipped every guardrail stepped by a post_call pipeline, even when the pipeline is left out of the stream for lacking the unified apply_guardrail interface, so a default_on guardrail that only implements async_post_call_streaming_hook stopped governing streams it governed on the merge base. The skip set now comes from the pipelines that gate the stream, the same way the iterator hook already computes it |
||
|---|---|---|
| .. | ||
| helpers | ||
| prisma_and_spend | ||
| proxy_logging | ||
| __init__.py | ||