x402 is the right base layer for machine-native payments. It’s here, and as of 04/02, it’s part of the Linux Foundation.

But it ships without the one thing every production system needs: a key and session story. That’s the layer to build.

What is x402?

x402 is an open protocol that builds on top of a more than twenty year old IETF standard: HTTP’s 402 Payment Required status code.

x402 is good because it is boring in the right way. It is simple, neutral, and very easy to explain.

The whitepaper’s pitch is straightforward:

Enable machine-native processes (e.g. AI agents, Web Services, etc.) to autonomously pay for API access, data, and digital services by leveraging the HTTP 402 “Payment Required” and internet-native payment rails (e.g. stablecoins, blockchains, etc.).

Having spent years deep in the crypto industry, one of the most important lessons I learnt is that simple protocols get adopted, while overdesigned protocols get discussed. That is the tradeoff. In my opinion, it’s why Bitcoin took off. It remains one of the most approachable whitepapers I’ve ever read. Just re-read it a handful of times and you’ll understand it.

How does x402 work?

The full x402 story is very easy to explain:

1. A client asks for a resource; such as a PDF.
2. The server replies 402 Payment Required; including a price and a destination address for payment.
3. The client pays; along with the crypto digital equivalent of a signed check.
4. The server returns the resource; after validating and setting the check.

That is the perfect base layer. It can’t be any simpler.

However, in the same way that the OSI Model needed a mature and ever-evolving Application Layer on top of the Transport Layer, x402 will need a mature and ever-evolving Key Management Layer on top of the Payment Layer.

The last page of the x402 whitepaper touts the following benefits:

• Instant, low-cost transactions.
• No API keys, no subscriptions, no middlemen.
• AI-first, developer-friendly, and blockchain-agnostic.

So what is x402 Missing?

Something I’ve learnt in the world of software development is that rewrites almost never work. You have to meet the world where it is, integrate with existing systems, and incrementally build towards the future you want to see. That is the only way to get from “this works on paper” to “this works in production.”

It’s been decades, and we still haven’t reached 50% IPv6 adoption as of writing this post.

If you want this to work in real systems, you need to account for how existing systems work, including things like:

• Scoped access keys
• Sessions keys
• Rotation and revocation of keys
• Admin key management

Then, the moment you enter the world of crypto, you learn that “custody hot keys” is a setup for a nightmare. You start to learn about multi-sig, hardware wallets, ring signatures, and we haven’t even gotten started with smart contracts.

What does it look like?

The missing key to the puzzle is not a change to the protocol, but building infrastructure on top of it that makes it usable with where the world is today, not where it’s going to be.

In fact, I really want the protocol to stay small and simple. It enables the ecosystem to introduce small, composable, layered primitives.

What’s the takeaway? In addition to enabling permissionless purchase of resources on the internet, enabling permissionless purchase of API keys is the next logical step. It is the missing piece that turns a one-off payment into an ongoing relationship with an existing service.

It doesn’t feel as “pure”, but it works. Keys are not a cop-out. They are the mechanism that turns a payment event into an ongoing service relationship.

This enables lots of different paradigms:

• Authenticate: Establish identity once per session instead of re-signing a payment on every call.
• Authorize: Scope each key to the minimum access it needs.
• Rotate/Revoke: Kill compromised keys in seconds without touching the protocol.
• Delegate: Hand scoped, short-lived keys to sub-agents without exposing the root wallet.
• Meter & Throttle: Attribute usage, enforce quotas, and flag abuse using existing observability and rate-limiting stacks.
• Integrate: Plug payments and top-ups into existing balance and billing systems.

Buying keys with x402 doesn’t have to be limited to API Keys. This could extend to session keys, license keys, support JWTs, or even just a more traditional pay-per-use model. The point is that the key management layer is what turns the protocol from a one-off payment into an ongoing relationship with a service.

Those are the boring parts that make a system shippable. And boring is good here.

How are others solving it?

First, what is the core problem of using off-the-shelf x402 for real systems? Hyperscale.

When you’re servicing millions of requests per second, you can’t go through a signature ceremony with every request. There’s a long list of problems to solve including, but not limited to, rate limiting requests, verifying payments, preventing abuse, protecting against denial-of-service, etc.

Two of the biggest efforts in the space targeting this problem are:

Circle has a very simple solution built on top of x402: batch thousands of nano transactions into one x402 transaction. It’s not a new protocol, it’s a Gateway based approach to solving this problem.

Stripe has a more ambitious solution, also built on top of x402: MPP. It gets real things right, most notably sessions: “A streaming payment over a payment channel. Use for pay-as-you-go or per-token billing with sub-cent costs and sub-millisecond latency.” Payment channels for machine-to-machine billing are a genuinely useful primitive.

The question isn’t whether those primitives are needed. They are. It’s whether they belong in a protocol, or in an SDK on top of one.

MPP bundles them with a new protocol, first-class support for its own blockchain, and its card network:

In the same way Cloudflare shipped Pay Per Crawl as a product and not a protocol, a lot of MPP could live as an SDK on top of x402. That’s the unbundling I’d want to see.

x402 feels pure and neutral, like Bitcoin, Ethereum, or Zcash. MPP solves real problems, but bundles them in a way that blurs the line between an open protocol and a product offering.

If you’ve been in crypto long enough, you’ve seen this pattern: a vendor ships an SDK that quietly becomes the de facto protocol, and composability pays the price.

Where do we go from here?

We find a delicate balance between building, waiting, and seeing what is a real problem that needs solving.

Stablecoins are here to stay. Agents will become more prevalent on the internet. The real work is distinguishing signal from noise about where the market actually is today. Big players can afford to fund speculative bets long past the point a thesis pays off, the way Meta has with the Metaverse. Smaller companies have to follow the customer that comes back.

At Grove, we shipped a wallet-first Linktree that is ready on the API and skills front for when the agents arrive. When we started, our thesis was that agents will pay humans. In reality, if you dive deep into x402scan.com or agentic.market, you’ll find a lot of tinkering, but not a market yet. If we had visibility into the numbers behind projects.dev, I’m sure we’d find the same thing.

To summarize, I’m extremely bullish on everything above. x402 will proliferate. MPP will be a great product. Agents will be consuming resources on the internet. I simply want to call out that in order to move the needle, we need to introduce layers that integrate the old system with the new, because it’ll be a while. That is the missing key to x402.

The open question: does the key layer belong in x402 itself, in an SDK ecosystem around it, or in a product like MPP? I believe the middle path wins.

If you’re building on top of x402, I’d love to hear what you think. In addition, not instead.

Originally published by Daniel Olshansky at https://grove.city on April 24, 2026.

The Missing Key to x402 was originally published in Grove on Medium, where people are continuing the conversation by highlighting and responding to this story.

What if you could get a creator you follow online to notice you?
What if you engage with your fans and earn from them directly?
What if all of this were powered by a simple, frictionless experience with low fees?

We’ve been working closely with our partners at Coinbase Developer Platform to integrate Apple Pay directly into Grove.

What’s missing from the flow today?

Most people are happy to support someone when the moment is right.

You see a cool reel on Instagram from a creator you like, and you want to show your appreciation in that moment. Maybe you just want to send them $0.02 for their thought.

Today…

• 🌐 You go to their Linktree
• 💳 You find their Patreon page
• 🔐 You sign up, or spend 5 minutes looking for your password
• 🧾 You enter your credit card information, again
• 🤔 You debate if you’re willing to commit to another monthly subscription
• ⏳ You lose the impulse to support them, and move on

What’s the issue?

Friction kills small acts of support.

Your Grove profile page exists to make it easy for others to support your original content creation.

It’s not just a set of links.
It doesn’t force anyone into a commitment.
It is a way for you to engage with your fans and earn from them directly.

It is your wallet-first landing page.
Your content.
Your fans.
Your money.
All in one place.

What Grove does differently

Grove puts payment rails directly on your profile page, alongside your content.
That means a fan can:

• 📲 Open your Grove profile
• 🍎 Fund their wallet with Apple Pay
• 💸 Tip you for a piece of content they saw online

Why does Apple Pay matter?
It reduces the friction to getting started.

What does this unlock?

When signing up and funding is easy, the profile becomes much more than a landing page.

1. Tip your content

If someone likes your work, they can support you on the spot.

They can paste in the link to your content or just tip you for being you.

2. Tip to Talk

If someone wants your attention, they can pay to reach you.

That makes the trade explicit: they pay for your attention in your inbox.

Read more about TipToTalk here.

3. Giveaways for your fans

You can run a giveaway to engage with your fans. All through the same experience.

4. QR Codes for Digital & Physical Businesses

This is not only for digital creators, it’s also great for physical businesses.

You can create a Grove page for your business, print out the QR code, put it at the front desk or next to the register, and let your customers support you by opening up the camera on their phone.

That turns a static page into a storefront.

What’s next?

Reach out to our team on Discord to learn more about how you can use Grove

Note: Apple Pay is only available in the US and requires you to connect an email address and verify a phone number. We support funding your tipping wallet through, as well as other crypto wallets.

Originally published by Daniel Olshansky at https://grove.city on April 16, 2026.

Grove partners with Coinbase Developer Platform to bring Apple Pay to Content Creator Earnings was originally published in Grove on Medium, where people are continuing the conversation by highlighting and responding to this story.

Your home mirrors you. You invite people over to share your space, your interests, your life.

Your personal landing page should be no different.

Linktree is like a parking lot. People come, only to leave as soon as they get what they need.

Grove is your personal landing page. People come, they stay, they support you, and they come back when they want to reach you again.

It should do more than just hold a list of links:

• ✅ Prove you’re a real person
• 💸 Turn your attention into revenue
• 🎁 Let people support you instantly
• 💬 Let people talk to you
• 🧠 Show what you care about
• 🏠 Feel like a homepage, not a parking lot

That’s what Grove is for 🌿

What it looks like

It’s a shift from a flat list of links to a home for your content, support, and money.

Why the old model is not enough

Linktree solved a real problem:

“I need one place to put my links.”

That worked when the internet was mostly about sending people somewhere else.

Now the page itself matters.

People are not just clicking through.

They’re deciding whether to trust you, follow you, pay you, or talk to you.

They’re deciding whether your attention is worth their time and money, and vice versa.

What Grove does differently

Grove does two things a plain link page can’t.

1. It puts money front and center 🪪

Grove is wallet-first.

Money isn’t hidden behind a redirect.

It’s part of the profile.

That unlocks a few things immediately:

• 💸 Tips — someone likes your work, they can pay you on the spot.
• ✉️ Messages — if someone wants your attention, they can pay to reach you.
• ⚡ Instant settlement — no waiting, no thresholds, no mystery balance somewhere else, no egregious fees.

2. It verifies who you are across multiple platforms ✅

Anyone can paste a URL.

That doesn’t prove ownership.

Grove lets you connect and verify accounts like X, GitHub, YouTube and more, so people know they’re looking at the real you.

In a world of impersonation and AI-generated everything, that’s not a feature. It’s the baseline.

Tip to Talk is the clearest example

We built Tip to Talk because attention has real monetary value.

If you want my attention, pay for it.

That sounds blunt. It’s also the point.

We all get hit with spam, cold DMs, “quick questions” that are never quick, and people asking for free time, free advice, or free introductions.

Tip to Talk makes the trade explicit:

• If you need someone’s attention, pay to reach them.
• If someone wants yours, they can pay to earn it.

That keeps the inbox open without making it free.

“Isn’t this just Patreon?”

Not really. Patreon solved recurring support.

Grove is built for direct, lightweight, just-in-time action.

That means:

• 😌 No subscription fatigue
• 🙏 No commitment just to say thanks
• ⚡ No extra friction for a $5 moment

Most people are happy to pay when they want to show appreciation. They just don’t want five more recurring subscriptions.

It’s awkward to not tip in person. It’s also awkward to cancel a recurring Patreon subscription.

Grove lets you send a message, show appreciation, instantly, and move on. That’s it.

Where do I start?

Linktree is a list of links. 🔗🪾

Grove is a digital home that verifies you and pays you. 🌿

If you already have a Linktree, move over. If you don’t, sign up anyway.

Check it out at Grove.city.

What comes next? Discovery, agent support, and better ways for people to find and pay creators. The goal is simple: make Grove the best way to present yourself online with modern technology that actually serves you.

Originally published by Daniel Olshansky at https://grove.city on April 8, 2026.

The missing feature in Linktree was originally published in Grove on Medium, where people are continuing the conversation by highlighting and responding to this story.

Your inbox is full. Your DMs are overflowing. Agents are spamming your replies. Everyone wants a piece of your attention, and none of them are paying for it.

The problem with open DMs

If you’re a creator, influencer, or business owner, you’ve been here before: dozens of cold messages, spam, and “quick questions” that aren’t quick. Everyone wants a coffee chat. The signal gets buried under noise.

Most platforms give you two options: open DMs to everyone, or close them entirely.

Open means chaos. Closed means missed opportunities.

Neither is great.

Introducing Tip to Talk

Today, we’re launching Tip to Talk on Grove.city:

A simple way for creators to get paid for their attention.

Here’s how it works:

1. Creators set a minimum tip to receive a message (e.g. $5)
2. Supporters tip and attach a message in one step
3. The creator gets paid, and the supporter gets seen

That’s it. No subscriptions. No monthly fees. Instant settlement.

Is this a new idea?

X announced limiting replies to “friends of friends”.

Y Combinator had an old trick where founders would send a VC money on Venmo to get their attention.

The desire to pay for someone’s attention isn’t new — it just hasn’t been easy.

“Take a human desire, preferably one that has been around for a really long time… identify that desire and use modern technology to take out steps”
- Ev Williams

At Grove, we’re building this in as a core feature. Tipping to get a message across is a fundamental primitive of a wallet-first, agent-friendly Linktree.

How to set it up

If you already have a Grove account:

1. Go to Dashboard → Settings → Engage
2. Toggle on Tip to Talk
3. Set your minimum amount
4. Done. Your profile now shows “Open to Tip to Talk”

It takes about 10 seconds.

Why this matters — get paid for your attention

Whether it’s a human or an AI agent, if someone wants your attention, they should pay for it.

💬 For creators — Your attention is your most valuable asset. Tip to Talk lets you put a price on it. Set the minimum at $1 or $100 — you’re in control.

🎯 For supporters — Want to guarantee a creator actually sees your message? Put your money where your mouth is. A $5 tip cuts through the noise better than a cold DM ever will.

🤖 For agents — AI agents can now pay to communicate with creators directly. If an agent finds your content valuable enough to reach out, it pays the price of admission like everyone else. No special API, no backdoor — just a tip.

Start earning with Grove 🌿

Create your account at Grove.city, enable Tip to Talk, and share your profile. We’d love to hear how you use it.

Follow us on X for more updates.

Originally published by Daniel Olshansky at https://grove.city on March 27, 2026.

Tip to Talk: Getting Paid For Your Attention Is All You Need was originally published in Grove on Medium, where people are continuing the conversation by highlighting and responding to this story.

When it comes to digital content, reach used to be king. Then revenue entered the conversation.

Companies like Substack and Patreon made revenue and subscriptions a first-class primitive. Companies like Medium followed, and Twitch followed. Now we’re seeing YouTube, Reddit, X, and others experiment with different models of integrated monetization.

Now, with AI training on and inferring from content, credit and compensation are front and center.

The opportunity is not about squeezing more out of the same revenue model. It’s about unlocking new ones.

The future of content publishing needs configurable revenue models, not just policy toggles

Medium’s latest update frames the problem well around 3Cs and their AI strategy: Consent, Credit, and C ompensation.

The current implementation is a binary toggle. Maximize reach or minimize training. Writers choose between getting cited by AI products, having AI learn about their contributions, or opting out entirely.

That is a step in the right direction. But it is not the 3Cs. It is a tradeoff between two of them.

Digg’s recent reset looks like a different story, but it is not. If bots, spam, and synthetic engagement dominate the system, trust disappears and the product breaks. Same root issue.

@nikitabier of X called this out near the beginning of February:

And the team at X did not wait to start acting on it. Nikita followed up a month later, saying the financial incentive to spam on X will soon be negative.

The internet still runs on economics that assume human attention is scarce, content is static, and abuse is manageable.

None of that is true anymore.

• 🗑️ Slop leads to a low-quality internet
• 🔇 Noise diffuses our attention
• 🤖 Abuse extends beyond keyboard warriors
• 🎭 Engagement is cheap to fake
• 🚧 Distribution is increasingly intermediated
• 🪞 Attention is increasingly fragmented

Luckily, the tools to solve this problem is already here: agentic payment infrastructure.

The infrastructure is already here, just not evenly adopted.

Not Everyone Wants the Same Thing

We started working on Grove.city late last year to make sure that content creators can engage their audience and earn what they deserve for high-quality content. It doesn’t matter if the creator or consumer is a human or agent: high quality earns what it deserves.

Speaking with various content creators, we learnt that not everyone wants the same thing:

• Some want to engage their audience
• Some want to make a living from it
• Some are building a personal brand
• Some just do it as a hobby

All of this results in tradeoffs that compete with each other: monetization, exclusivity, exposure, reach, distribution, quality, engagement, attention, and more.

But the current landscape forces everyone into the same set of tradeoffs:

• Want reach? You give up revenue.
• Want revenue? You gate content and lose reach.
• Want engagement? You optimize for algorithms that reward volume over quality.

The one thing we know for sure is that no one says no to a solution that simultaneously maximizes Reach and Revenue without compromising on content quality.

Business Models: The Reach vs. Revenue Landscape

Here is the current landscape of business models on the Reach vs Revenue tradeoff matrix, including new models enabled by agentic payment infrastructure:

It’s not about giving up one revenue stream for another. It’s about having more ways to earn. It’s about increasing optionality. It’s about growing the pie for everyone.

What Options Do We Have?

Below is a non-exhaustive list of options companies like Medium could implement to continue enabling reach, create revenue opportunities, and account for AI access.

1. Pay to unlock

One article. One payment. No subscription.

Most paywalls require a monthly subscription. But what if you could pay to unlock just a single article or video for a small fee?

This creates a middle ground the web mostly skipped over: not free, not subscription, just pay for what you want, when you want it.

2. Pay for preview

Bots pay to peek. Humans decide if it’s worth it.

You’ve likely seen a handful of articles where you get a preview (a few paragraphs) for free, but then you have to pay to read the rest.

Paying to unlock solves for humans, but not for bots. What if you had to pay a couple of cents to see a preview?

Any bot or AI lab training their models would have to pay to see if they even want to train on your content. Then, they choose if they want to convert from paying for a preview to paying for the full article or not.

It might not amount to much, but it’s a different revenue model that ensures the creator gets something for their work.

Put your money where your mouth is.

Up until now, comments have been a binary toggle: open to everyone or completely off. Most platforms turned off comments a decade ago. That was the only available answer.

What if there was a middle ground? Pay to comment.

Are you a keyboard warrior who is willing to give $0.05 to share your frustration? If you’re willing to pay, go for it.

Are you a supporter who wants to express your gratitude and get the author’s attention? A small economic attribution goes further than a cold email.

Are you a bot that’s spamming the internet? Well, if you’re willing to pay, maybe some people would be okay with it.

If posting is free, spam scales. If commenting requires a small payment or stake, low-effort abuse becomes more expensive. Quality goes up. Moderation gets easier. Community trust improves.

4. Leave a tip

Let agents and humans say “thank you” with real value.

We’ve all heard of Sentiment Analysis, but how about Sentiment Creation?

Imagine if Frontier Labs gave their agents a tipping jar — leaving a small tip during training or inference whenever they deemed content useful. I’d probably tweet about it every time it happened.

This has been tried many times, but it never really took off because there is no sustainable long-term business model for it. What we’ve seen is that this won’t survive as a standalone product but can live side-by-side with the business models above by leveraging the same infrastructure.

Is the infrastructure for this ready today?

Yes.

Protocols, standards, infrastructure, usage. It’s all here. Simple, cheap, and available to everyone. Just hasn’t been widely adopted yet.

The building blocks exist. What’s missing is adoption.

A 🌿 from Grove

If you’re interested in integrating a new revenue model into your are not sure where to start, we’re here to help. Send We’ve got a lot of exciting product updates coming soon. Follow us on or create an account at to get early access. us a message and we’ll help you out.

Originally published by Daniel Olshansky at https://grove.city on March 25, 2026.

An Open Letter to Content Publishers: New Ways to Earn was originally published in Grove on Medium, where people are continuing the conversation by highlighting and responding to this story.

When we first launched Grove, we supported one platform: X.

The goal was simple:

Let people tip creators anywhere on the internet.

But creators don’t exist in just one place.

You might write on Substack, post on X, upload to YouTube, stream on Twitch, contribute on GitHub, or publish on your own website.

Your identity — and your work — lives across the entire internet.

That’s why we introduced Creator Profiles on Grove.

Profiles allow creators to link all of their platforms together into a single identity, so fans can support them no matter where their content lives.

If you missed that launch, you can read more about how Grove lets you tip creators across every platform on the internet in our earlier post:
👉 https://grove.city/news/profiles-are-live-tip-creators-on-every-platform

Today, Grove supports over a dozen social networks and creator graphs — from X to YouTube to personal websites and beyond.

But expanding across the internet introduced an important challenge:

How do we make sure the right person gets paid?

The Problem

If fans are sending money across the internet, they need to know one thing:

That the creator they’re supporting is actually… the creator.

🚫 No impersonators.
🚫 No fake accounts.
🚫 No middlemen.

Just direct support from fan → creator.

That’s why we built Profile Verification.

Introducing Profile Verification

Profile Verification ensures that every account attached to your Grove profile actually belongs to you.

When you sign up on Grove you can:

1. Create your Grove profile
2. Attach your social accounts
3. Verify ownership using secure authentication (OAuth)

Once verified, those accounts become part of your portable creator identity.

Fans can see exactly where you exist online and know that they’re sending tips to the real you.

Why This Matters

For creators

• Your identity becomes unified across platforms
• Your reputation travels with you
• Your fans know where to support you

For supporters

• You know your payments are going to the right person
• No scams or impostors
• Just direct support for the creators you love

Verification protects both sides of the ecosystem.

This is how the internet should work.

Start Verifying Your Profile

• Claim your handle
• Connect your social accounts
• Verify ownership
• Start receiving tips across the internet

Setting up your verified creator profile takes just a few minutes.

What’s Next

Verification is just the first step.

• A unified creator identity
• A reputation graph across platforms
• And a world where creators can earn anywhere on the internet

It lays the foundation for a much bigger vision:

value can flow directly to the people who create it.

Because when identity is verified…

🌿

Originally published by fredt at https://grove.city on March 24, 2026.

Your Identity Matters. Introducing Profile Verification. was originally published in Grove on Medium, where people are continuing the conversation by highlighting and responding to this story.

Streamlining code reviews for a content earnings platform with agents.

P.S. I took the images on my widescreen monitor while working on a real feature. If there’s interest, I’m happy to put together smaller screenshots or a walkthrough video.

TL;DR

Our workflow today looks roughly like this:

• Kick off an agent review locally via /grove_*_review
• The agent reviews the diff, runs best-practice checks, and spins up the system locally
• Full E2E flows run automatically (happy, sad, chaotic paths)
• Failures are diagnosed with context instead of just “test failed”
• A PR sweep catches regressions and tech-debt risks
• Humans review mission-critical logic manually
• A final PR description is generated automatically

The key idea: Leverage tools off the shelf, semi (not completely) automate things specific to your domain, and do as much of it locally as possible.

Code Reviews are still the Bottleneck

It seems like everyone is talking about how code reviews are the new bottleneck in the era of agentic software development.

There is some truth to it, but if you’ve been around long enough, you know it’s not new. The problem just looks a little different now. The approach and the solutions are evolving. They’re not the same, but they rhyme.

There are lots of approaches, tools, and companies tackling this problem.

We’ve tried or looked into Claude CodeReview, CodeRabbit, and PropelCode. They’re all good and will get you at least halfway there.

It’s the part that’s specific to your domain, your product, your tech stack, your culture, and your taste.

However, the other half is the hard one.

Most importantly, it meets developers where they already are.

This is just a quick show-and-tell of our lightweight workflow around Claude Code. It’s simple, custom-tailored, and easy to use.

The Stack

Three repos:

The majority of these codebases were developed in an agent-first environment.

The code in the frontend and extension is what some would refer to as “vibe-coded”.

The backend is a bit more mature and reviewed in depth because it deals with fund management. I still review the core logic line-by-line, but I haven’t written a single line of it myself.

CI vs Local Automation

We still run traditional CI, but the role has changed.

Traditional CI only runs linting and unit tests.

The heavier work happens locally and integrates (not replaces) the human. The agent spins up the stack, runs E2E flows, and reviews the branch before a PR even exists.

This approach is similar to the direction described by DHH when moving CI back to developer machines.

Our end-to-end tests also double as production smoke tests that run on GitHub after deployment. 🔥

Step 1: Kick off the review via /grove_*_review

Each repo has a custom local slash command that kicks off the review:

• runs git diff against the default branch
• builds context for the agent
• checks cosmetic changes
• validates best practices
• prepares E2E testing

The command does a handful of things:

Let’s assume you just engineered (not just “vibe-coded”) a big new feature with the help of agents, and it’s time to start reviewing your work.

Documentation is a living thing. It’s the responsibility of both the human and the agent to update and review these commands and files regularly whenever something new is learned, a pattern emerges, or a change happens.

‼️ Like any AGENTS.md file, this is not one-and-done. I use /session-commit to keep it updated.

Step 2: DO ALL THE THINGS

Once the initial review completes, it returns a report, a summary, and a list of actionables.

• updating docs or flows
• fixing terminology
• adding unit/integration/e2e tests
• adding logs
• following existing patterns
• linting

The actionables cover many things we’ve learned we need to manage:

Running end-to-end tests is one of the most important steps.

Usually I tell it to DO ALL THE THINGS, but it really depends on the situation.

Step 3: Review the results

Here the human actually reviews the results and jumps in where needed.

In this particular case, Docker wasn’t running, so the end-to-end tests couldn’t start. Not a big deal, but something I prefer not to delegate to an agent.

Step 4: E2E Test Results

This is one of my favorite parts of the API.

It spins up a database, starts the server, and runs realistic end-to-end flows against it.

Happy paths, sad paths, chaotic paths. 🎢

Everything.

If something fails, the agent tries to diagnose the root cause before surfacing it.

No “test failed” without context. It tells you why. 🔍

Claude also fixes some of them (if trivial) along the way, and is instructed not to fix anything where the business logic change is questionable or requires another opinion.

The E2E tests report back with a clear pass/fail matrix.

Step 5: Pull Request Sweep

From here, I use my personal agent skills before moving into manual review.

In particular, I’ve found that cmd-pr-sweep from Olshansk/agent-skills is great at catching major regressions, bugs, and limiting how much tech debt we're taking on.

If you’re curious, the skills can be installed with:

npx add skills olshansk/agent-skills

I review it manually and decide what to fix.

Having that inline is a great way to give future agents context about the tech debt.

If something feels too big or out of scope for the work, I ask the agents to add TODOs with a lot of context.

This is the last gate before a human looks at the PR. 🚧

Step 6: Prepare and upload the PR

Once everything passes, I commit, push, and generate a PR description using /cmd-pr-description.

The skill reads the full diff, commit history, and review findings, then drafts a structured description with a summary, feature diff table, and technical details.

Here is what it looks like on GitHub:

Other Special Mentions

• Cross-referencing with other models Depending on the size and complexity, I like to cross-reference plans and bugs with Gemini and Codex. Gemini tends to be very idiomatic and strong at frontend work. Codex is great at architecture and challenging requirements. Not every change needs this, but for architectural decisions or tricky edge cases, it’s useful to get a second or third opinion from another model.
• Manual review I don’t review frontend code manually, but I still review mission-critical business logic line-by-line on GitHub. I leave comments (locally or remotely) and have agents pick them up. I focus heavily on reducing code surface area, regression testing important edge cases, naming, and ensuring TODOs with explanations are in place.
• /session-commit I use this frequently to keep AGENTS.md updated based on learnings from the most recent agent session.

This isn’t intended to be a “how I code” post, but I wanted to call out some of the patterns I’ve found useful lately:

Depending on how much work happened during this stage of the review, I might go back and start the process again.

Originally published by Daniel Olshansky at https://grove.city on March 13, 2026.

What Semi-Automated Code Reviews with Claude Code Look Like at Grove was originally published in Grove on Medium, where people are continuing the conversation by highlighting and responding to this story.

We just shipped our biggest update yet. Creators now have public profiles, and your audience can tip you everywhere — X, YouTube, TikTok, and more. Here’s what changed.

Your Profile Is Your Home Base

Every Grove creator now has a shareable profile at “ grove.city/<username>” (here’s mine: grove.city/arthur).

Profiles pull together everything in one place:

• your connected social accounts
• your earning wallet
• your lifetime stats

Think of it as your new creator landing page, but one that actually lets people pay you. Anyone can tip you directly from your profile. Share your link in your post, or video description, or drop it in your newsletter. That’s it.

Giveaways, Referrals, and the Leaderboard

Grove isn’t just about collecting tips. There’s a whole layer of gamification built in to help you grow.

Keep more of what you earn.

• You keep 90% of every tip.
• Grove takes a 10% fee to cover its cost.

Earn when your fellow creators earn

Referrals earn you passive income.

• Every account gets a referral link. When someone you refer starts earning tips, you get 2% of their tip volume for a full year.
• Refer other creators, share your link, and earn while they earn.

Engage with your fans

Giveaways let you engage your audience with time-gated events that your followers can enter.

• Set the minimum tip amount to count as an entry
• Set the length, and Grove picks the winners.
• It’s a simple way to reward your community and generate additional revenue.

Climb to the top of the charts.

The Leaderboard tracks it all. See who’s tipping, who’s earning, and where you stand, with live tips, top tippers, and top earners sections. It’s public, it’s competitive, and it gives your audience another reason to support you.

What’s Next?

We’re making your profile the center of your experience on Grove, which means we’ll be surfacing more useful information there for your and your fans, like your active giveaways, your most popular/paid posts, and more!

Claim your profile at grove.city and start earning today.

Originally published at https://grove.city on March 5, 2026.

Profiles Are Live: Tip Creators on Every Platform was originally published in Grove on Medium, where people are continuing the conversation by highlighting and responding to this story.

One of a16z’s recent posts, 8 big ideas for 2026, touched on a topic that stuck with me: the invisible tax on the open web.

Table of Contents

Two problems collide

The core idea is simple and uncomfortable.

“AI agents extract data from ad-supported sites (the context layer), providing convenience to users while systematically bypassing the revenue streams (like ads and subscriptions) that fund the content.”

And the only real path forward is a system that:

“could include models like next-generation sponsored content, micro-attribution systems, or other novel funding models. Existing AI licensing deals are also proving to be a financially unsustainable Band-Aid, often compensating content providers with a fraction of the revenue they’ve already lost to AI-cannibalized traffic.”

This reminded me of an observation Ilya Sutskever (ex Chief Scientist of OpenAI) made in his 2024 NeurIPS talk. At minute 8, he states that data, the fossil fuel of AI, is not growing!

Those two problems collide.

If original human content becomes economically irrational to produce, and AI systems depend on fresh, diverse, ground-truth data, the system eats itself.

So the question becomes: How do we fund original content on the open web without rebuilding paywalls, killing distribution, or forcing creators into platform lock-in?

And how do we do it in a way that works for both humans and agents?

An old but new primitive

Our answer is micro-tipping.

Not tipping as charity. Not tipping as guilt. Not tipping as a replacement for subscriptions, payments or paywalls.

Tipping is a lightweight, optional economic signal that creates alignment.

It uses AI, stablecoin payment rails, and leverages incentives that operate at both human and machine scales. It does not require accounts, contracts, or commitments. And it works everywhere.

We are already building this at Grove. We are early, but you can see the direction if you take a look.

Why would humans tip humans?

There are two reasons:

1. It feels good
2. It’s a lottery ticket to an experience

Tipping feels good when it is optional and low-stakes. The person sending the tip feels good, as does the person receiving it. If you micro-tip two cents at a time, you can feel good dozens of times a day.

Large payments feel like investments. Subscriptions feel like commitments. They come with expectations, pressure, deliberations, and second thoughts. Small payments feel expressive. They say, “I liked this,” without asking for anything in return.

Most importantly, this doubles by enabling a missing incentive platform that connects supporters to creators, irrespective of the platform they’re on.

Every tip can double as a lottery ticket.

Leveraging stablecoin rails uses a public ledger. Grove, or anyone else, can build a platform or leaderboard where creators can connect with their audience.

For example, if you make “A life in the day of” videos, you’re tipping your way into a chance to spend a day in the life of the video creator. If it’s an apparel maker, you’re tipping your way into sweepstakes for that free pair of sneakers.

This is true interoperability. Anyone can see who tipped whom, what was at stake, and who won. It is internet native, platform-agnostic, and AI-first.

A small chance at a meaningful upside, access, recognition, or connection will dramatically affect human behaviour. One cent is no longer trivial if it is one more entry. Leaderboards work. Scarcity works. Humans are predictable this way.

Generosity alone does not scale. Incentives do.

Why would agents tip humans?

Synthetic data is important, but it is not enough. Derivative content helps learning, but it has its limitations.

Ground truth still has to be produced manually, by expert humans, in the real world, with taste, context, and judgment. Frontier models already know this. They pay for data today through brittle licensing deals and private pipelines.

Original free online human-created content is not an OR, it’s an AND.

Micro-tipping turns this into an open market.

Agents do not need dopamine. They need access. They need freshness. They need diversity. Tipping becomes a clean, programmatic way to compensate creators for marginal value without negotiating contracts or enforcing paywalls.

Until the singularity shows up, original human content remains a dependency.

This is not about paywalls

New protocols like x402 have their place in the future of the web, but they’re not an endgame in themselves.

Paywalls have their place. Free content has its place. Micro-tipping lives in the middle.

It is optional. It does not block access. It does not fragment the web. It does not force creators to choose a platform.

Today, every major social platform is reinventing tipping in isolation. YouTube, Twitch, X, Reddit, etc. There is no composible layer to view or act on this data, but the infrastructure for the payment rails already exists, they’re simply unused.

What we are missing is a universal balance (a tipping jar) that humans and agents alike can use anywhere. Imagine a Venmo-like primitive for the open web, for every social graph, with the same user experience wherever you go.

Call it micro-tipping. Call it micro-attribution. Call it micro-support. Call it a micro-donation. The important thing is that we’re not paying micro-soft.

The name matters less than the behaviour.

Tipping already exists as a social construct. We tip drivers, performers, and waiters. Whatever your opinion of tipping culture, it is not going away. It is being redefined in the AI era.

The open web needs a native way to say “this mattered” without closing itself off.

That is the story we are telling. And we think it unlocks a win for creators, users, agents, and the internet itself.

Originally published by Daniel Olshansky at https://grove.city on February 10, 2026.

Why Micro-Tipping Is the Fuel for Original Content on the Open Web was originally published in Grove on Medium, where people are continuing the conversation by highlighting and responding to this story.

When I stepped into the role of CEO last year, I doubled down on our mission at that time:

• Deliver the Shannon upgrade to Pocket Network, enabling permissionless demand, generalized data transport, and automated mint/burn tokenomics.
• Ship PATH, the reference implementation for a Pocket Network Gateway, based on our own battle-tested gateway.
• Launch new verticals on Pocket Network.

On June 3, 2025, we launched the Shannon upgrade. On November 4, 2025, we reached the 1.0.0 milestone for PATH, enabling anyone to launch their own permissionless gateway. As for scaling to new verticals, teams like POKT.ai are already building those integrations today.

With these accomplishments, Grove has made the decision to step away from directly operating or developing the protocol and our gateway, the Grove Portal. Throughout the month of November, the vast majority of Grove’s paid and public endpoint traffic will migrate to new public endpoints infrastructure hosted on https://api.pocket.network, ensuring the Pocket community continues to have the reliable access it needs. All of our remaining RPC and Validator infrastructure has moved to a new home: BuildInTheShade.com, which catalogues all of our past and present RPC-based partnerships and initiatives. This marks an important milestone: Grove is moving from Gateway to Application, from infrastructure provider to product builder.

Why Agents and Payments

For the past two years, Grove has been exploring the intersection of agents and payments, looking at how crypto and modern payment rails together can enable new coordination between humans, software, and machines. AI agents are here, and they need the ability to compensate the humans, creators, and services they depend on.

We have shared this vision publicly in talks and writing:

• AI Agents Are Coming (YouTube)
• AI Agents and the XRPL-EVM (YouTube)
• AI Agents and Payments (LinkedIn)

What We Are Building

Our next step is to make this vision concrete with a new product, The Grove API.

The Grove API is a simple, but powerful payments API that allows Humans and AI (agents, assistants, browsers, enterprises etc…) to tip, compensate, or pay humans and content platforms with a single API call. Behind the scenes, Grove will handle custody, routing, compliance, attribution, and automatic currency conversion, so developers and agents can focus on what matters.

Key features include:

• One API call to pay any creator or platform
• Support for stablecoins initially, with a fast follow for crypto, and fiat with automatic conversion
• Enterprise-grade attribution tracking, risk management, and audit trails
• Built on emerging standards such as x402 for agent micropayments and MCP for agent-to-service interoperability
• Support for attribution metadata (LLMs.txt, AI.txt, JSON-LD, etc.) so agents know who to pay automatically
• Designed for humans to tip anyone anywhere on the web for anything.
• Designed for AI use cases like training data compensation, inference usage, and rewarding content on the open web
• More on the larger vision can be found here: https://buildwith.grove.city

The mission is clear: to enable new forms of value transfer in the emerging world of crypto and AI, while giving both humans and agents a practical path to compensating creators and services.

What Happens Next

To make the transition smooth, here is what you can expect:

• Pocket Network’s public endpoints and related pages including docs, the status page, and chain pages have moved to https://api.pocket.network.
• Paid traffic will be sponsored by the Pocket Network Foundation during the migration period, and for a time afterward, through these new public endpoints.
• Grove’s customers have already been notified to make the transition in their apps, and have begun doing so since November 1, 2025.

To everyone who has trusted Grove over the years — developers, partners, and node runners — thank you! This pivot is not an ending. It is the continuation of the Pocket story, moving farther up the stack to where applications meet the end user.

Stay tuned. We are excited to share what comes next.

Grove’s Next Chapter: Moving Up the Stack was originally published in Grove on Medium, where people are continuing the conversation by highlighting and responding to this story.