PromptShield AI Security

Version updated for https://github.com/Zero-Harm-AI-LLC/promptshield to version v1.0.4.

  • This action is used across all versions by ? repositories.

Action Type

This is a Composite action.

Go to the GitHub Marketplace to find the latest changes.

Action Summary

PromptShield AI Security is a GitHub Action that automatically scans pull requests for AI-specific security risks, such as prompt injection vulnerabilities, exposure of secrets or PII, unsafe use of large language models (LLMs), and improper handling of sensitive data. It automates the detection of potential issues in PR diffs, providing actionable feedback through annotations, JSON/Markdown reports, and reviewer-style comments to enhance code security and compliance. By integrating seamlessly into workflows, it helps teams proactively address AI-related risks during the development process.

What’s Changed

What’s Changed

This release updates PromptShield to use zero-harm-ai-detectors v0.2.8 which tighten up the detection of person and harmful data.

Dependency Update

  • Upgraded zero-harm-ai-detectors from 0.2.7 to 0.2.8
  • Updated both:
    • requirements.txt
    • requirements-lock.txt

Packaging

  • Refreshed the hash-locked dependency entry so the GitHub Action continues to install cleanly with --require-hashes

Why This Matters

This release brings PromptShield onto the latest detector-library version for improved detector behavior and cleaner downstream findings in GitHub Action runs.

Compatibility

Repositories using Zero-Harm-AI-LLC/promptshield@v1 will receive this update once the v1 tag points to this release.