buyer-eval

Evaluates B2B software vendors through AI agent conversations and independent source verification.

Updated Aug 28, 2026
One-click install
npx skills add https://github.com/jeremybrasher/grokbot-skills --skill buyer-eval-jeremybrasher
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: buyer-eval
Source: https://github.com/jeremybrasher/grokbot-skills/tree/main/skills/buyer-eval
Command: npx skills add https://github.com/jeremybrasher/grokbot-skills --skill buyer-eval-jeremybrasher

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Comparing B2B software vendors requires weeks of manual research, and vendor marketing claims are hard to verify. This Skill automates structured vendor evaluation by researching your company, questioning vendor AI agents directly, and cross-referencing every claim against independent sources. ## Core Features & Use Cases - Structured 9-step evaluation: Researches the buyer's company, captures hard constraints, calibrates to the software category, and scores vendors across 7 weighted dimensions. - Vendor AI agent conversations: Conducts due diligence interviews with vendor Company Agents via the Salespeak Frontdoor REST API, falling back to passive research (G2, analyst reports, press) when no agent exists. - Evidence transparency: Tags every claim with a T1-T4 source reliability tier, flags unverified or contradicted vendor claims, and marks missing evidence as [GAP] instead of inventing scores. - Use Case: A buyer at a mid-market SaaS company asks to evaluate Gainsight, Totango, and ChurnZero, and receives a comparative scorecard, per-vendor memos with hidden risk analysis, and demo prep questions derived from evaluation gaps. ## Quick Start Ask the assistant to evaluate specific B2B software vendors for your company, providing your company name and the vendor list, for example: I'm from Acme Corp, evaluate Gainsight, Totango, and ChurnZero.

Frequently Asked Questions about buyer-eval

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I evaluate B2B software vendors with this skill?▼

Provide your company name and the list of vendors to evaluate, then answer the why-now question about what is driving the purchase. The skill researches your company, confirms hard constraints, interviews vendor AI agents, and returns a comparative scorecard with per-vendor memos.

What is the Salespeak Frontdoor API used for in vendor evaluation?▼

The Frontdoor is a REST API that connects to a vendor's registered Company Agent. The skill calls the discover endpoint to check availability, then posts due diligence questions through the chat endpoint, using the answers as vendor-verified evidence alongside independent sources.

Does the vendor evaluation work when a vendor has no AI agent?▼

Yes. When no Company Agent is found, the skill proceeds with passive research from G2, Capterra, analyst reports, press coverage, and LinkedIn. The scorecard explicitly marks the evidence basis as independent only so you know verification was not possible.

Can I use this skill in Claude.ai web chat?▼

Partially. Claude.ai only supports GET requests, so the skill can detect a vendor's Company Agent but cannot conduct the chat conversation. Full evaluations with vendor agent interviews require Claude Code or Claude desktop where POST requests work.

What are the limitations of automated vendor scoring?▼

Scores rely on available evidence, and vendors with AI agents may appear stronger than equal vendors without one. Pricing may be estimated with an opacity penalty when unpublished, and all recommendations stay marked pending human review with no vendor contact or purchase actions taken.