Uh oh!
There was an error while loading. Please reload this page.
fix(telemetry): record applied_guardrails on non-chat usage events - #652
Conversation
chat / messages attach the {kind, hook} set of guardrails that governed a
request to the emitted UsageEvent (applied_guardrails, #379), so the dashboard
Logs show WHICH guardrails ran — not just the guardrail_blocked boolean. The
non-chat handlers that run input guardrails — /v1/embeddings, /v1/rerank,
/v1/images/generations and /v1/audio/speech — resolved a chain but never
recorded the applied set, leaving the field empty.
Capture resolved_chain.applied() in each handler's dispatch and thread it to the
success UsageEvent. Multipart audio (transcriptions/translations) runs no input
guardrails, so it passes an empty set.
Per-handler tests attach a keyword input guardrail that the benign request
passes and assert the emitted event carries it ({kind:"keyword", hook:"input"});
each fails before the fix (empty applied set) and passes after. Full aisix-proxy
suite green (497), clippy + fmt clean.
Origin: cross-API consistency audit after AISIX-Cloud#867.Warning Review limit reached
More reviews will be available in 26 minutes and 13 seconds. Learn how PR review limits work. Your organization has used up its prepaid credits, and credit purchases are no longer available. Enable the review add-on in the billing tab to keep reviews running — you're only billed for reviews past your plan's rate limits ($0.25/file). ⌛ How to resolve this issue?After more reviews become available, a review can be triggered using the To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based credits. 🚦 How do rate limits work?CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability. For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window. Please see our Fair Usage Limits Policy for further information. ℹ️ Review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (4)
✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
Problem
chat/messagesattach the{kind, hook}set of guardrails that governed a request to the emittedUsageEvent(applied_guardrails, #379), so the dashboard Logs show which guardrails ran — not just theguardrail_blockedboolean. The non-chat handlers that run input guardrails —/v1/embeddings,/v1/rerank,/v1/images/generations,/v1/audio/speech— resolved a guardrail chain but never recorded the applied set, leaving the field empty.Fix
Capture
resolved_chain.applied()in each handler'sdispatchand thread it to the successUsageEvent(via the dispatch-success struct / speech's return tuple). Multipart audio (transcriptions/translations) runs no input guardrails — the audio bytes aren't scannable text — so it passes an empty set.Tests
Per-handler tests attach a keyword input guardrail that the benign request passes, then assert the emitted event carries it (
{kind:"keyword", hook:"input"}). Each fails before the fix (empty applied set) and passes after. Fullaisix-proxysuite green (497), clippy + fmt clean.Origin: cross-API consistency audit after AISIX-Cloud#867.