AIDevToolsTestingSaaS

Agent Usability Scanner

Test your product from an AI agent's point of view and surface where agents get stuck

Problem
More users reach products through AI agents, but developers can't see where agents stall, how often they retry, or how many tokens they waste. Existing analytics only track human behavior and miss agent experience.
Solution
A CLI that runs simulated agents through key user journeys (signup, checkout, search) and records blockers, retries, and token spend, producing an agent-usability report with concrete fixes like simpler forms or machine-readable hints.
Users
SaaS teams and developer-tool authors serving AI-agent traffic
MVP
CLI tool; 3 preset agent task templates; record step latency and retries; generate Markdown reports; support custom task scripts
Revenue
Open-source CLI free; hosted reports and historical comparison at $19/mo
Why now
Agent-driven product usage is emerging, early attempts exist but no standard tool; indies can iterate fast

Source discussions · 2

  1. Hacker NewsShow HN: I asked HN about AI agentic setups, here are 750 of the comments mapped4↑ 0
  2. Hacker NewsShow HN: AJX – Find out where AI agents get stuck using your product1↑ 0
← Back to list