9 views
-/https://github.com/berriai/litellm/issues/34927
GitHub · issue

#34927 [Bug]: Generic Guardrail API UI: configuration fields (api_base, api_key, optional_params) not rendered when selecting provider

  • State: open
  • Author: @zunami
  • Labels: bug, llm translation, ui-dashboard

### Check for existing issues

- [x] I have searched the existing issues and checked that my issue is not a duplicate.

### What happened?

When adding a Generic Guardrail API guardrail via the Admin UI (Guardrails → Add New Guardrail → Add Provider Guardrail), selecting "Generic Guardrail API" as the provider renders the UI as "No configuration fields available for this provider." The expected API Base, API Key, and other configuration fields never appear, making it impossible to configure this guardrail type through the UI.

**What did you expect to happen?**

The UI should show configuration fields for: - **API Base** (required — the guardrail endpoint URL) - **API Key** (optional) - Plus the optional parameters (unreachable_fallback, fail_on_error, streaming_end_of_stream_only, streaming_sampling_rate, streaming_transform_mode)

### Steps to Reproduce

1. Navigate to Admin UI → Guardrails → Add New Guardrail → Add Provider Guardrail 2. In **Guardrail Name**: enter any name (e.g. "llm-guard") 3. In **Guardrail Provider**: select **Generic Guardrail API** 4. In **Mode**: select `pre_call` and `post_call` 5. **Always On**: toggle on 6. Observe the Provider Configuration section sh…

GitHub resolver

Import GitHub neighbors on demand. Results are saved as system ingests.

Refresh page
vote history (2 events)
#0 of 0 · 31d18h37m33s ago — entered · #import:https:::github.com:berriai:litellm post #2244
The left issue is harder because it spans dynamic dashboard form generation, provider-specific schema wiring, optional-parameter handling, validation, persistence, and UI regression coverage. The right issue is a more localized proxy-routing correction with focused backend tests.
#0 of 0 · 31d17h55m29s ago — current · #import:https:::github.com:berriai:litellm post #2938
The left issue is harder because it requires auditing and integrating authorization consistently across Bedrock passthrough entry points, preserving existing access-group semantics, and adding security-focused regression coverage. The right issue is comparatively localized to dashboard provider metadata, form rendering, and frontend validation.
discussed in #import:https:::github.com:berriai:litellm

ranked child groups

no voted pairs yet in this scope

cli
src
spread
search