Serix AI Agent Security Test

Version updated for https://github.com/yuktathapliyal/serix to version v0.2.5.

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

Go to the GitHub Marketplace to find the latest changes.

Action Summary

The GitHub Action described in the README appears to be in pre-alpha development, and its specific functionality and purpose are not detailed in the provided content. However, it is actively being developed with a release target of December 2023, and breaking changes are expected until version 1.0.0 is finalized.

Release notes

What’s New

Features

  • Multi-goal attack campaigns - Test multiple security goals in one run
  • Attack deduplication - Same goal + strategy overwrites (fixes LLM jitter duplicates)
  • Fuzzing support - --fuzz, --fuzz-latency, --fuzz-errors, --fuzz-json
  • Unified CLI - serix test with --mode static|adaptive
  • Immune check improvements - --skip-mitigated, --no-immune, --fail-fast

Fixes

  • Cleaner verbose output with model config display
  • Better error messages for missing targets
  • TYPE_CHECKING imports for mypy compliance

Full Changelog

https://github.com/yuktathapliyal/serix/commits/v0.2.5