Buyer-eval is an open-source Claude code/cowork skill that evaluates B2B software vendors the way a senior analyst would, except it takes 30 minutes, not 3 weeks, and it's completely free. Give it your stack, budget, and shortlist. It talks to vendor AI agents, cross-references every claim against G2 and analyst reports, and scores across 7 weighted dimensions with full evidence tracking. Walk into every demo knowing exactly where to push. MIT licensed. No API key needed.

Progress Agentic RAG — Trusted AI answers with citations, deployed in days

Trusted AI answers with citations, deployed in days

Hey PH, I'm Yarin, and I run growth at Salespeak AI. We built buyer-eval after watching the same pattern repeat: buyers spend weeks researching, sit through six demos, and still walk in without the right questions. The vendor who presents best wins. Not the one who fits best. The finding that came out of running it surprised us: across countless evaluations, vendors with an AI agent on their website won 91% of head-to-heads. The methodology is fully transparent, it's not a bias in the tool. Verified, real-time evidence simply scores higher than claims that can only be cross-referenced against G2 or press. That gap is widening. Everything is open, the tool, the methodology, and all 27 results. Run your first eval in minutes, no API key needed. Happy to answer any questions about how the scoring works or what we've seen across categories.

Progress Agentic RAG — Trusted AI answers with citations, deployed in days

Trusted AI answers with citations, deployed in days

Buyer-eval

Stop trusting vendor demos. Interrogate them with AI first.

Stop trusting vendor demos. Interrogate them with AI first.