openai/openai-guardrails-pythonPublic

NotificationsYou must be signed in to change notification settings
Fork17
Star126

Dev/steven/remove anonymizer#47

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to ourterms of service andprivacy statement. We’ll occasionally send you account related emails.

Already on GitHub?Sign in to your account

Jump to bottom

Merged

gabor-openai merged 5 commits intomainfromdev/steven/remove_anonymizer

Nov 10, 2025

Merged

Dev/steven/remove anonymizer#47

gabor-openai merged 5 commits intomainfromdev/steven/remove_anonymizer

Nov 10, 2025

+492 −44

Conversation

Copy link

Collaborator

steven10a commentedNov 10, 2025•
edited
Loading

Remove dependency onpresidio-anonymizer to prevent package clash

presidio-anonymizer is being used to mask detected entities. This PR implements that functionality ourselves without depending on an external library
Additionally, updatesBIC_SWIFT detection which was resulting in too many false positives
Updated and added relevant tests

steven10a added3 commits

November 10, 2025 14:44

Remove dependency on Presidio Anonymizer

2b0d899

Mask in sorted order

695b0d4

Improve overlap detection

bc611a8

steven10a requested review fromCopilot andgabor-openai

November 10, 2025 20:01

CopilotAI reviewed

Nov 10, 2025

View reviewed changes

Copy link

CopilotAI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others.Learn more.

Pull Request Overview

This PR removes the dependency onpresidio-anonymizer and implements custom text masking functionality to prevent package conflicts. The changes include a new lightweight anonymizer module, improved BIC/SWIFT detection patterns to reduce false positives, and comprehensive test coverage for both the anonymizer and BIC detection.

Key changes:

Implemented custom anonymizer insrc/guardrails/utils/anonymizer.py to replacepresidio-anonymizer
Enhanced BIC/SWIFT detection with context-aware patterns and bank code whitelisting to reduce false positives
Added baseline tests to ensure the custom anonymizer matches expected behavior

Reviewed Changes

Copilot reviewed 6 out of 6 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
src/guardrails/utils/anonymizer.py	New module implementing custom PII masking functionality with overlap resolution
src/guardrails/checks/text/pii.py	Updated to use custom anonymizer and improved BIC/SWIFT detection patterns
src/guardrails/_base_client.py	Migrated from`presidio-anonymizer` to custom anonymizer for structured content masking
tests/unit/checks/test_pii.py	Added tests for BIC/SWIFT detection including false positive prevention
tests/unit/checks/test_anonymizer_baseline.py	New baseline tests ensuring custom anonymizer produces expected results
pyproject.toml	Removed`presidio-anonymizer` dependency

💡Add Copilot custom instructions for smarter, more guided reviews.Learn how to get started.

src/guardrails/checks/text/pii.py OutdatedShow resolvedHide resolved

Fix known bank codes

cd2173e

steven10a requested a review fromCopilot

November 10, 2025 20:12

Copilotstarted reviewing on behalf ofsteven10a

November 10, 2025 20:12

View session

Copilot finished reviewing on behalf ofsteven10a

November 10, 2025 20:13

CopilotAI reviewed

Nov 10, 2025

View reviewed changes

Copy link

CopilotAI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others.Learn more.

Pull Request Overview

Copilot reviewed 6 out of 6 changed files in this pull request and generated 10 comments.

💡Add Copilot custom instructions for smarter, more guided reviews.Learn how to get started.

src/guardrails/_base_client.py OutdatedShow resolvedHide resolved

tests/unit/checks/test_anonymizer_baseline.py OutdatedShow resolvedHide resolved

Clean up test files

5ffd47d

steven10a requested a review fromCopilot

November 10, 2025 20:22

Copilotstarted reviewing on behalf ofsteven10a

November 10, 2025 20:23

View session

Copilot finished reviewing on behalf ofsteven10a

November 10, 2025 20:26

CopilotAI reviewed

Nov 10, 2025

View reviewed changes

Copy link

CopilotAI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others.Learn more.

Pull Request Overview

Copilot reviewed 6 out of 6 changed files in this pull request and generated 3 comments.

💡Add Copilot custom instructions for smarter, more guided reviews.Learn how to get started.

src/guardrails/checks/text/pii.py

		"ANZB\|NATA\|WPAC\|CTBA\|"# Australia
		"BKCH\|MHCB\|BOTK\|"# Japan
		"ICBK\|ABOC\|PCBC\|"# China
		"HSBC\|SCBL\|"# Hong Kong

Copy link

CopilotAINov 10, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others.Learn more.

"HSBC" is duplicated in the known_bank_codes string. It already appears in line 176 as part of the "Major international" banks. Consider removing the duplicate "HSBC" from the Hong Kong section to avoid redundancy.

Suggested change

	"HSBC\|SCBL\|"# Hong Kong
	"SCBL\|"# Hong Kong

Copilot uses AI. Check for mistakes.

Copy link

CollaboratorAuthor

steven10aNov 10, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others.Learn more.

[nit]

tests/unit/checks/test_anonymizer_baseline.py

Comment on lines +162 to +165

		# Check that at least the first format is detected
		assert"<PHONE_NUMBER>"inresult1.info["checked_text"]# noqa: S101

Copy link

CopilotAINov 10, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others.Learn more.

Thetexts_and_results list is created and appended to but never used. Consider removing this unused variable or adding assertions that use it.

Suggested change

	# Check that at least the first format is detected
	assert"<PHONE_NUMBER>"inresult1.info["checked_text"]# noqa: S101
	# Check that all formats are detected
	fororiginal,maskedintexts_and_results:
	assert"<PHONE_NUMBER>"inmasked# noqa: S101

Copilot uses AI. Check for mistakes.

Copy link

CollaboratorAuthor

steven10aNov 10, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others.Learn more.

This is a fine unit test for now

src/guardrails/checks/text/pii.py

		"ICBK\|ABOC\|PCBC\|"# China
		"HSBC\|SCBL\|"# Hong Kong
		"DBSS\|OCBC\|UOVB\|"# Singapore
		"CZNB\|SHBK\|KOEX\|HVBK\|NACF\|IBKO\|KODB\|HNBN\|CITI"# South Korea

Copy link

CopilotAINov 10, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others.Learn more.

"CITI" is duplicated in the known_bank_codes string. It already appears in line 176 as part of the "Major international" banks. Consider removing the duplicate "CITI" from the South Korea section to avoid redundancy.

Suggested change

	"CZNB\|SHBK\|KOEX\|HVBK\|NACF\|IBKO\|KODB\|HNBN\|CITI"# South Korea
	"CZNB\|SHBK\|KOEX\|HVBK\|NACF\|IBKO\|KODB\|HNBN"# South Korea

Copilot uses AI. Check for mistakes.

Copy link

CollaboratorAuthor

steven10aNov 10, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others.Learn more.

[nit]

gabor-openai approved these changes

Nov 10, 2025

View reviewed changes

gabor-openai merged commitbf65130 intomain

Nov 10, 2025

9 checks passed

gabor-openai deleted the dev/steven/remove_anonymizer branch

November 10, 2025 20:30

Labels

None yet

		"ANZB\|NATA\|WPAC\|CTBA\|"# Australia
		"BKCH\|MHCB\|BOTK\|"# Japan
		"ICBK\|ABOC\|PCBC\|"# China
		"HSBC\|SCBL\|"# Hong Kong

		"ICBK\|ABOC\|PCBC\|"# China
		"HSBC\|SCBL\|"# Hong Kong
		"DBSS\|OCBC\|UOVB\|"# Singapore
		"CZNB\|SHBK\|KOEX\|HVBK\|NACF\|IBKO\|KODB\|HNBN\|CITI"# South Korea

Movatterモバイル変換

Dev/steven/remove anonymizer#47

Dev/steven/remove anonymizer#47

Conversation

steven10a commentedNov 10, 2025• editedLoading Uh oh!There was an error while loading.Please reload this page.

Uh oh!

Uh oh!

CopilotAI left a comment

Choose a reason for hiding this comment

Pull Request Overview

Reviewed Changes

Uh oh!

Uh oh!

CopilotAI left a comment

Choose a reason for hiding this comment

Pull Request Overview

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

CopilotAI left a comment

Choose a reason for hiding this comment

Pull Request Overview

Uh oh!

CopilotAINov 10, 2025

Choose a reason for hiding this comment

Uh oh!

steven10aNov 10, 2025

Choose a reason for hiding this comment

Uh oh!

CopilotAINov 10, 2025

Choose a reason for hiding this comment

Uh oh!

steven10aNov 10, 2025

Choose a reason for hiding this comment

Uh oh!

CopilotAINov 10, 2025

Choose a reason for hiding this comment

Uh oh!

steven10aNov 10, 2025

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

steven10a commentedNov 10, 2025•
edited
Loading