Agentprobe

What is Agentprobe?

Agentprobe is a web checker for merchants who want AI shopping agents to be able to buy from them. You submit a seller URL and it runs 13 probes. They look for MCP support and commerce.json, the discovery files llms.txt, OpenAPI and ai-plugin.json, the product catalog, quote and checkout endpoints, payment rails, refund and contact details, a declaration of unsupported rails, fake claims and proof of fulfillment. The result is a score out of 110 and a grade: CERTIFIED at 85 or higher, AGENT_READY from 70 to 84, PARTIAL from 40 to 69 and NOT_READY below 40. You can run it through a probe API, an MCP tool called probe_site, an embeddable badge, or a GitHub Action that gates a CI pipeline. A public registry lists every probed site and its grade, with 37 sites listed and 14 of them certified when this was written. The product is described as a free probe API, and no paid plans are published.

Use cases

• Test a store for AI agent readiness
• Check llms.txt and MCP availability
• Audit catalog, quote and checkout APIs
• Add a readiness badge to a site
• Gate deployments with a CI check
• Compare a site against certified sellers

Who it's for

• Merchants
• Web developers
• E-commerce teams

Why use Agentprobe

• You get a specific list of missing signals, so you know what to build before agent shoppers arrive.
• Grade tiers make it clear whether a site is discoverable, transactable or opaque to agents.
• The badge and registry let you show a passing score publicly.
• The CI action catches regressions in agent-facing endpoints on every change.

Agentprobe features

• 13 readiness probes per seller URL
• Score out of 110 with four grade tiers
• Probe API by GET or POST
• MCP endpoint with a probe_site tool
• SVG badge for any probed domain
• Public registry and JSON leaderboard
• GitHub Action for CI checks
• Checks for payment rails, refund and contact info

Agentprobe integrations

• GitHub Actions

How to get started with Agentprobe

1. Open agentprobe.fly.dev and enter the URL of the seller site to test.
2. Agentprobe runs its 13 probes against that site.
3. Read the score and grade, and see which probes failed.
4. Fix the gaps, such as adding llms.txt or an OpenAPI file, and run the probe again.
5. Optionally add the badge or the GitHub Action to track readiness over time.

Agentprobe limitations

• Measures machine-readable readiness, not actual agent purchases
• No paid plans or usage limits are published on the site

Agentprobe pricing

Agentprobe is free to use.

Agentprobe plans

• Probe API

  Free

  • Probe any seller URL
  • MCP probe_site tool
  • Badge and registry listing

Plans as listed on the official pricing page on 24 Oct 2026. Prices can change and may vary by country or exclude tax. Check the official pricing page.

Agentprobe alternatives

The main Agentprobe alternatives listed on Debuly are SandBase Harness, SDKProof, skill-optimizer, Agent Governance Toolkit, Flowise and Marqo. The table compares pricing and platforms; order is A–Z.

1. 

   Agent Governance Toolkit

   Microsoft's open-source toolkit for enforcing policy, identity, sandboxing and audit records on autonomous AI agents before they reach production.

   Developer platformsAutomation and agents
   Open sourceVisit Agent Governance Toolkit (official site)
2. 

   Flowise

   Open-source visual builder for AI agents, chatbots and RAG workflows, self-hosted or on Flowise Cloud.

   Developer platformsAutomation and agentsChat assistants
   Open sourceVisit Flowise (official site)
3. 

   Marqo

   AI-native ecommerce search and product discovery platform that learns from shopper behavior and your catalog.

   Developer platformsMarketing and SEO
   Contact salesVisit Marqo (official site)
4. 

   SandBase Harness

   Open-source, local-first runtime for AI agents with sandboxed tools, persistent sessions, memory, credential vaults, audit trails and a console.

   Developer platformsAutomation and agents
   Open sourceVisit SandBase Harness (official site)
5. 

   SDKProof

   Command-line tool that checks which library APIs AI coding assistants get wrong, using the TypeScript compiler to judge each answer.

   Developer platformsCoding
   FreeVisit SDKProof (official site)
6. 

   skill-optimizer

   Docker workbench and agent skill that runs repeatable, graded evals on agent skills across OpenRouter models.

   Developer platformsCodingAutomation and agents
   Open sourceVisit skill-optimizer (official site)

Agentprobe FAQ

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

• Logo: none collected yet, so a monogram is shown.
• Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.