Skip to content
Beseam
Commerce Fieldbook
Projects and references/Agent skills and protocols

Promptfoo

A framework for evaluating prompts, agents, and model behavior with repeatable test cases and assertions.

Open-source projectStableReviewed 2026-07-25
Open official source

How the Fieldbook uses it

We refer to Promptfoo in work involving agent skills and protocols. Its own documentation remains the authority for installation, compatibility, and current behavior. A listing here does not imply endorsement by its maintainers.

Good fit for

  • AI answer regression checks
  • Commerce-agent workflow evaluation
AI evaluationGEOAgents

Before you adopt it

Maintainer
Promptfoo and contributors
License or usage boundary
MIT

Check the official source for the current release, support policy, compatibility notes, and license terms. The review date on this page only records when Beseam last checked the listing.

Related guides

problem · emerging

Our products do not appear in AI shopping answers

Work out whether the gap comes from the questions being tested, product data, source visibility, or the answer system's own selection.

Open workflow

skill · stable

Ecommerce technical SEO and GEO audit

Check crawl access, rendered content, product markup, merchant facts, and answer-source visibility without collapsing them into one score.

Open workflow

skill · emerging

AI citation-gap investigation

Compare a fixed set of buying questions, cited pages, merchant attribution, product facts, and competing products.

Open workflow

playbook · emerging

AI citation investigation playbook

Coordinate a repeatable review when shopping-answer systems omit products, cite competitors, or state the wrong facts.

Open workflow
Also in Agent skills and protocolsAnthropic Agent SkillsAlso in Agent skills and protocolsAgent Skills specificationAlso in Agent skills and protocolsShopify AI ToolkitAlso in Agent skills and protocolsModel Context Protocol