agentprobe-injection is an open-source CLI tool for evaluating the resistance of LLM agents to indirect prompt injection attacks. It allows researchers and developers to benchmark and compare the effectiveness of different defense strategies, supporting AI safety and robustness efforts.
agentprobe-injection is a LLM eval & observability product. It enables developers to test and compare LLM agent resistance to indirect prompt injection attacks. It is built as an open-source project for AI safety researchers and LLM developers. agentprobe-injection is open source under the MIT license. agentprobe-injection is available on the command line.
agentprobe-injection first shipped in 2026. PulseGate's similarity index finds few close equivalents — agentprobe-injection occupies a relatively distinct niche. Key capabilities include prompt injection testing, LLM agent evaluation, and security benchmarking.
Latest indexed changes and source events
agentprobe-injection discovered by the PulseGate indexer
Other apps tracked under the same category.
