sentinel-ai-auditor is a Python package that performs deterministic, offline security auditing of AI agent skills and instruction bundles. It uses static analysis to detect issues such as prompt injection without requiring live model inference. The tool outputs results in SARIF format and is designed for integration into developer workflows and CI/CD pipelines.
sentinel-ai-auditor sits in PulseGate's AI & LLM security category. It focuses on detecting security vulnerabilities like prompt injection in AI agent skills and instruction bundles without relying on live model calls. sentinel-ai-auditor is an open-source project aimed at AI developers and security researchers. sentinel-ai-auditor is open source under the MIT license. It runs on the command line.
sentinel-ai-auditor first shipped in 2026. Key capabilities include Static Analysis, offline-first, and SARIF Output.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do