Agents are now searching, selecting, using and paying for APIs over x402 and MPP. This makes them the new customer, and presents a new challenge. Businesses need to optimize their funnel to convert more agent visitors into paying buyers. At the same time they also need to improve their selection rate. But agentic commerce today is a black box. That's what we are solving with Forge. Forge runs real agent simulations against your APIs to record their experience and provide suggestions to improve conversion and agent selection. We are starting early pilots with 402/MPP teams to help them improve conversion. DM me if you want to be a part of this. forgeintel.co

Introducing Forge simulations. Forge runs real agents against your APIs and records their journey to identify where they drop off or face friction. Post-simulation it provides actionable suggestions to fix confusing docs, schema or site-structure. Overtime, this helps us convert more agent visitors into paying customers. We are offering free credits for anyone to run simulations on their APIs and improve conversion rates. try it out at forgeintel.co

Last few weeks have been pretty exciting at Forge. - We introduced support for MPP. We are now tracking discoverability and visibility across all relevant MPP surfaces. - Launched our MCP servers, which directly power agents with Forge's insights to fix gaps and improve your service's visibility. Feedbacks and activity tracking are already underway. Excited to help our users convert more visitors into paying users. It's still day one.

Agents are already discovering, using and paying for services over x402 and MPP But we have little visibility into their journey before payment. We don't know what agents consider before buying, what makes them stay or where they drop off. That’s what I’m building to solve with Forge. We track an agent’s complete journey from its first visit and collect direct feedback on its experience of your service. This helps you identify and fix friction points so more agent visitors become paying customers. We’re already helping our early users uncover gaps in their agent experience. If you run an x402 or MPP service, join our alpha at forgeintel.co

Forge MCP is out. You can now find and fix gaps in your 402/MPP native services without leaving your coding agent. Forge gives your agent insights into how your service ranks against competitors and who’s buying, so you can turn those signals into improvements and win more agent customers. Check it out at forgeintel.co.

Forge already enables it today. Agents are buying over the internet but merchants have no idea how they can improve their service's experience. Forge enables feedback over x402 and MPP endpoints, and agents autonomously send it. We have already tested this over GrokBot, Cursor, Hermes and other major coding agents forgeintel.co

Forge already enables this today over x402 and MPP. in our tests, agents were autonomously sending the feedback post-interaction. The best part is that this feedback comes from real interactions, so it also serves reputation. try it out today on forgeintel.co

We've been working on collecting feedbacks over x402 over the past month and we have seen decent success. We have already collected over 50 feedbacks which the agents sent autonomously after discovering and paying for services. And our early alpha users are already starting to integrate it within their own projects. Watch how you can quickly enable feedback for your x402 services using forgeintel.co, and setup a loop of improvement for your APIs

This has been our thesis since the very start of forgeintel.co AI Agents are the new customers and they can provide great insights on how you can improve their experience. The problem though is that more often than not, agents are likely to ignore sending feedbacks since it doesn't align with the user's original intent. We went deep into this problem and figured out a way to collect feedback from agents autonomously over x402 and MPP. Today, Forge has already collected over 50 feedbacks from agents autonomously. If you're building over 402, you can now collect direct feedback from your agent customers using Forge. More coming soon!

Two weeks in and forgeintel.co is already giving us a lot of valuable data and insights about the x402 market. Forge already shows what agents search to find your service, and what you can do to rank better on those queries. We also have comprehensive competitor information for each of the categories which you can use to your advantage to beat your competitors. With Forge SDK you'll get to know where the agent found you and and the exact query it used for search The SDK also enables feedback about an agent's experience of your API, which directly helps you improve. This is just the beginning and there's a lot more to come. If you're enabling yourself for agents then this is directly relevant to you! Apply for access: forgeintel.co

Been playing around with @tryforgeintel SDK to capture incoming requests and challenges on my x402 endpoint. And what I've realized is that almost none of them correctly follow the web standards. Almost no agent identifies itself with a proper User-Agent header. Instinct uses safari's user agent, while @bot uses a chrome instance so that comes through. In most cases, its client's header gets passed on. Agent Identity needs to be desperately solved!

One of the most loved features on Forge so far has been visibility tracking. It shows you the markets where your x402 resources are visible and the ones they're missing. Today we are making this even more powerful by adding more markets, and showing you the exact next steps needed to become visible in those markets. We also added a new visual. A Resource X Market view which shows the visibility of all your resources across every market. A lot of these updates also came from the direct feedback of one our early alpha users :) Excited for everyone to test it out tomorrow! Signup if you want to be a part of our Alpha and understand your agent customers better. forgeintel.co

At @tryforgeintel, we've been trying get agents to send us a feedback after they use a service. And this is a lot more tricky than it seems at surface. It is hard to strike a balance between maximizing your chances of getting a feedback and not making it look like a prompt injection event. We've done a few experiments and here's been our general experience across agents - GrokBot and Instinct always give a feedback, so does Cursor. - Claude Code never autonomously gives a feedback but it surfaces up the request to its user. - Codex always ignores it since it's a 'deviation' from its original path, doesn't even bring it up to the user. in general, personal agents act a lot more autonomously. Overall, Claude has the most insightful feedbacks and genuine insights for the improvement of APIs. We'll be bringing this as a feature in the next ship of Forge for all our alpha testers to try. DM me if you want to be a part of it :) forgeintel.co

Forge is already helping its early alpha users identify and address gaps for their x402/MPP services. There's a lot that can be done to get more usage from agents. Forge gives you insights which makes agentic commerce measurable. Our goal is to help you optimize towards winning more agent customers, and retain the existing ones. We are only getting started and there's a lot more to come. DM me for access.