AgentLab is an open-source, local-first platform for testing AI agents with YAML suites, secure command execution, PASS/FAIL/ERROR analysis, and downloadable JSON/HTML reports. It includes a CLI and local web GUI with no cloud upload, database, or telemetry.
Hi Product Hunt! I built AgentLab because AI agents are difficult to trust without repeatable tests and clear evidence.
AgentLab lets developers define YAML test suites, run local agents securely, inspect PASS/FAIL/ERROR results, and export JSON/HTML reports—all locally, with no cloud upload or telemetry.
I’d love your feedback on the workflow and which evaluators you would like to see next.
Report
No reviews yetBe the first to leave a review for AgentLab