Agent claimed 'added to cart' with an empty cart — an offline trace-based testing platform catches it
SrinivasPad · reddit · 2026-11-22
Reddit user Srinivas, CTO/co-founder of testAIng.com, released AiEnsured agentest, an agentic AI testing platform born from a real failure: a shopping agent told the user the charger was added to the cart, but the recorded trace contained no addtocart call.
- Core idea: grade what the agent actually did from the recorded trace, not how well the reply reads.
- 165 tests run fully offline — no API key and no model in the pass/fail path, so the same trace always gets the same verdict.
- Install via pip install agentest-ai, run agentest demo or target the empty-cart case with agentest std run --agent a22 --ids T1-018 --format md; a 60-second Colab notebook is available plus PyPI, GitHub and a Hugging Face checklist.
More from coding & agent
- Browser dragon-in-your-hand demo built with Three.js, MediaPipe hand tracking and WebRTC — techartist_ · 2026-11-22
- Developer Praises Google Antigravity's Agent Swarm UX — majidmanzarpour · 2026-11-22
- Store Scraper MCP exposes App Store and Google Play data to AI agents — modelcontextprotocol · 2026-11-22
- etapa MCP offers a free AI cycling coach for beginners — modelcontextprotocol · 2026-11-22
- Claude Code plugin turns hooks into persistent memory shared across Codex and Cursor — blaizedsouza · 2026-11-22
- Brandon Galang: As Models Get Capable, UX Scaffolding Melts Away in Favor of Context Flow — brandongalang · 2026-11-22