Show HN: Claude skill that evaluates B2B vendors by talking to their AI agents
Evaluating B2B Vendors with AI: A New Approach
As someone who has been in the trenches of evaluating B2B software vendors, I can attest that the process has remained largely unchanged for decades. We've all been there - filling out lengthy forms, reading reviews on G2, and sitting through demos that seem designed to avoid our most pressing questions. But what if I told you there's a new way to evaluate vendors, one that leverages AI to get to the heart of what really matters?
How it Works
The Claude skill, built by the team at Salespeak, takes a different approach to evaluating B2B vendors. Here's how it works:
- It researches your company automatically, gathering information on your industry, size, and tech stack, so you don't have to fill out a form.
- It asks 2-4 category-specific questions to surface requirements you may not have known you had.
- It attempts to find and talk directly to each vendor's AI agent, using a REST API call to initiate a structured due diligence conversation.
- It asks tough, adversarial questions, such as "What are your customers' most common complaints?" and "What use cases are you NOT a good fit for?" - and flags when agents deflect instead of answering.
- It cross-references every vendor claim against independent sources, such as G2, Gartner, and press reports, in a Claims vs. Evidence table.
- It produces a scorecard with transparent evidence tracking, showing whether each score is backed by vendor-verified evidence or public sources only.
The Power of Agent-to-Agent Conversation
The agent-to-agent piece is what really sets this approach apart. When a vendor has an AI agent, Claude (working on behalf of the buyer) interrogates it directly, then fact-checks its answers. This allows for a much more nuanced and informed evaluation of each vendor. And when vendors have different evidence levels, the skill quantifies what would change if the missing evidence were confirmed, ensuring that no vendor is unfairly favored.
Installing the Skill
To get started with the buyer-eval skill, simply ask Claude to "Install the buyer-eval skill from salespeak-ai on GitHub." Then, use the /buyer-eval command to run it. The skill is MIT licensed, and the code is available on GitHub at https://github.com/salespeak-ai/buyer-eval-skill.
Who is this for?
This skill is designed for anyone who needs to evaluate B2B vendors, whether you're a small business owner looking for the right software solution or a large enterprise seeking to optimize your tech stack. With its transparent and evidence-based approach, this skill has the potential to save time and reduce risk in the vendor evaluation process.
So, what do you think? Are you ready to try a new approach to evaluating B2B vendors? What challenges have you faced in the past, and how do you think this skill could help? Share your thoughts in the comments below!