rulebench is an open-source CLI tool that allows AI agent developers to trap-test their agent rules defined in CLAUDE.md files. It runs isolated sessions to determine if rules are effective and provides honest delta reports. Ideal for developers working with agent frameworks and LLM-based systems.
In the LLM eval & observability space, rulebench takes a focused approach. It focuses on evaluating whether agent rules defined in CLAUDE.md files actually function as intended in isolated test sessions. It is built as an open-source project for AI agent developers. rulebench is open source under the MIT license. It runs on the command line.
rulebench first shipped in 2026. PulseGate's similarity index finds few close equivalents — rulebench occupies a relatively distinct niche. Key capabilities include rule testing, isolated sessions, and delta reporting.
Latest indexed changes and source events
Other apps tracked under the same category.
