fix(security): redact sensitive headers from debug logs#2918
Open
giulio-leone wants to merge 3 commits intoopenai:mainfrom
Open
fix(security): redact sensitive headers from debug logs#2918giulio-leone wants to merge 3 commits intoopenai:mainfrom
giulio-leone wants to merge 3 commits intoopenai:mainfrom
Conversation
At debug log level, response headers were logged unfiltered, potentially exposing API keys and authorization tokens. Now filters sensitive headers (authorization, api-key) from debug log output using the existing SENSITIVE_HEADERS set, consistent with SensitiveHeadersFilter. Fixes openai#1196
There was a problem hiding this comment.
Pull request overview
This PR addresses #1196 by preventing sensitive response headers (e.g., authorization, api-key) from being emitted in debug logs in the sync and async HTTP clients.
Changes:
- Redacts sensitive values when logging
httpx.Response.headersin bothSyncAPIClientandAsyncAPIClient. - Re-exports
SENSITIVE_HEADERSfromopenai._utilsto reuse the existing sensitive-header allowlist. - Adds a test intended to validate the response-header redaction logic.
Reviewed changes
Copilot reviewed 3 out of 3 changed files in this pull request and generated 3 comments.
| File | Description |
|---|---|
src/openai/_base_client.py |
Redacts sensitive response headers before emitting debug logs for HTTP responses. |
src/openai/_utils/__init__.py |
Exposes SENSITIVE_HEADERS for reuse outside _utils/_logs.py. |
tests/test_utils/test_logging.py |
Adds a test for header redaction behavior using SENSITIVE_HEADERS. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
…ase_client Adds a test using respx to verify that sensitive response headers are actually redacted in log output when making real client calls, not just testing the filter utility in isolation. Refs: openai#2918
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Fixes #1196
Response headers were logged unfiltered at debug level in both sync and async HTTP clients (
_base_client.py), potentially exposingauthorizationandapi-keyheader values in log output.Changes
src/openai/_base_client.py: Filterresponse.headersthrough a dict comprehension that redacts values for headers inSENSITIVE_HEADERSbefore logging. Applied to bothSyncAPIClientandAsyncAPIClient.src/openai/_utils/__init__.py: Export the existingSENSITIVE_HEADERSset from_logs.pyso it can be reused in_base_client.py.tests/test_utils/test_logging.py: Added test verifying response header redaction logic.Approach
Reuses the existing
SENSITIVE_HEADERSset ({"api-key", "authorization"}) from_utils/_logs.py, which is already used bySensitiveHeadersFilterfor request options. The filter only needed to cover request args in dict format; this PR extends coverage to response headers logged directly.Before
After