docs: update provider SDK auto-instrumentation coverage - #368
Conversation
Document AWS SDK v3 Bedrock Converse tracing for JavaScript. Describe default content capture and language-specific opt-out settings. Standardize supported API tables across provider guides.
|
Preview deployment for your docs. Learn more about Mintlify Previews.
💡 Tip: Enable Workflows to automatically generate PRs for you. |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 429c25d63f
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
| Provider prompt and response content is disabled by default because it can contain sensitive data. PromptLayer strongly recommends enabling content capture when your data policies allow it. Content capture is crucial for getting the full value from PromptLayer search and analytics because it provides the prompt, response, and tool data used by request inspection, content search, analysis, and debugging. | ||
|
|
||
| Without content capture, traces still include model names, timing, token usage when available, and other non-content telemetry. PromptLayer cannot search or analyze prompt, response, and tool content that was not exported. | ||
| PromptLayer captures supported provider prompt and response content by default. This provides the prompt, response, and tool data used by request inspection, content search, analytics, and debugging. |
There was a problem hiding this comment.
For users following the updated preload examples with the current SDKs, content is still not captured unless explicitly enabled: the JS SDK's resolveCaptureContent() returns false when OTEL_INSTRUMENTATION_GENAI_CAPTURE_MESSAGE_CONTENT is unset, and promptlayer/register calls configureTracing() without captureContent; Python configure_tracing() also does not set the OTel capture variable for the provider instrumentors. Removing the old span_only/true exports means these quickstarts produce metadata-only traces even though the docs promise prompt/response content, so please keep the enablement instructions or update the SDK defaults first.
Useful? React with 👍 / 👎.
2 participants
Summary
ConverseCommandtracing, setup, and content coverage.This aligns the docs with MagnivOrg/prompt-layer-js#195 and MagnivOrg/prompt-layer-library#342.
Validation
mint validatemint broken-linksgit diff --check