Launched this week
Running even one online store is a full-time job. SellerClaw is a team of AI agents that runs it for you: specialized agents for product sourcing, store management, and advertising, coordinated by a supervisor you direct. Tell it what to sell — the agents build listings, manage ads and pricing, and handle fulfillment and support across Shopify, eBay, and more. You stay in control: every action is visible and approvable, and you set how much runs on its own. Free to start.
Great stuff! Is it better for complete 0s or experienced sellers?
I was planning to start my first store, and I'm not sure if it's too much for me.
SellerClaw
@petropalovsk It can be great for both, but experienced sellers will usually get more value faster because they already have more workflow to unload onto the agent.
That said, it’s not “too much” for a first store — you can start simple and use it to take repetitive work off your plate as you learn.
SellerClaw
@petropalovsk Thanks! It actually works for both, but in different ways. If you're experienced, the agents take the repetitive grind off your plate. If you're starting from zero, they double as a head start, the Product Scout helps you find what to sell and the channel agents handle the listing and pricing mechanics you'd otherwise spend weeks learning.
So not too much for a first store. You set the guardrails, the agents do the heavy lifting, and you learn the business by watching what they do rather than drowning in it on day one. Plenty of first-time sellers stall on exactly the parts we automate.
If you go for it, I'm happy to point you to the simplest starting setup. Good luck with the first store either way.
The developer workflow angle is the interesting part for me. I might have missed it, but is there a way to keep changes reviewable before the AI applies them?
SellerClaw
@xiaosong001 Yes, the user can define the boundaries for what the agent is allowed to do autonomously, and for anything more sensitive you can set it up so changes go through manual review before they’re applied.
So you can run it in a more hands-off mode for low-risk tasks, or make it approval-based where every change stays reviewable first.
Mailwarm
I can see built with OpenClaw.
Does that mean you are built on top of it?
SellerClaw
@bengeekly Yes — OpenClaw is the core, but SellerClaw is much more than just a wrapper around it.
We use OpenClaw as the foundation and built a lot of customization around it for e-commerce: hosted cloud runtime, operational skills, integrations, automation logic, guardrails, and workflows that let it actually run store ops at scale. So “built with OpenClaw” is accurate, but the product itself is the specialized layer on top.
Your agent can integrate with Woocommerce
SellerClaw
@lovik1468
Not yet, WooCommerce isn't integrated at the moment. It's something we can look at adding soon, though. Right now our focus is on Shopify as the primary platform, alongside the leading marketplaces. Appreciate you flagging it, demand like this is exactly what helps us prioritize what comes next.
SellerClaw
@lovik1468 WooCommerce is definitely on our priority list for upcoming SellerClaw integrations. It’s one of the platforms we’d really like to add next.
Mailwarm
This is a big vision. Running a store touches so many moving parts: sourcing, listings, pricing, ads, fulfillment, support. Having specialized agents coordinated by a supervisor makes sense. The key challenge is probably trust and control.
What actions are merchants most comfortable automating first: listings, pricing, ads, or support?
SellerClaw
@thamibenjelloun Most start in advisory mode, which means the agent recommends and the merchant approves everything. What gets handed over first usually comes down to risk tolerance for that specific store. The modes exist precisely because that answer is different for everyone.
SellerClaw
@thamibenjelloun Hey,
What to automate first depends upon the business model and the platform you sell in.
1. The dropshippers need to automate listings and pricing model first since the new products on the supplier end appear every hour. It's a competitive advantage if you can list the product fast, set up a stock control to avoid OOS (out of stock) cancellations, add correct fees, shipping costs and your margin.
2. As to pricing, basically the US market is quite coherent in terms of marketplace fees. Amazon has a referral fee of 15% for most of the categories for years. Here's the top question is to calculate the economy of the listing not to lose your margin. Pricing matters but it's mostly a one-time set up and control of regular deals / discount you're ready to give.
3. Support is mostly applied to Etsy sellers as the platforms requires a day-to-day communicatios between the seller and the buyer and lots of offers have some customizations and made-to-order listings. Here's the agent can take control of it and in this case it's really looking for automation from scratch. The most crititcal here is to create a RAG model and provide an agent with the comprehenive description of the listings so the agent will find the information in your own database instead of googling the incorrect information which may cause a return and seller's costs.
4. Once all the ops questions closed you can keep going with ads. It's sure to be automated once the correct p&l dashboard with all the fees is in front of you and you clearly understand the ACOS & TACOS you can allow.
Best,
Gleb
Fundraisly
Nice launch! Curious how quickly credits get used in a real store.
SellerClaw
@solodnev
Thanks! Honest answer: real-world burn rate is something we're still gathering data on as more stores come online, so I won't give you a made-up number.
What I can tell you is how it works under the hood: different operations run on different models matched to the task. Routine, high-volume actions use lighter models, while the more analytical work runs on heavier ones. So credit usage scales with what your store actually does, its size, how much daily routine you automate, how much content gets generated, and so on, rather than a flat rate.
As we get more usage data, we'll be able to share real benchmarks. Great question to keep us honest on.
SellerClaw
@solodnev @tolstov_gleb
Adding to the above: the part you control directly is an effort slider that trades economy for depth, in three steps — Saver (lighter, faster models for routine work), Balanced (default: strong models on the hard tasks, light on the routine), and Max (strongest models across the board for the deepest research). So you set not just how much the agent does, but how thoroughly and how cheaply it does it.
SellerClaw
@solodnev Depends on the workflow. Active sourcing and listing across multiple channels uses more than a store running mainly support. We're on a credit-based model: three plans with larger monthly balances at higher tiers, plus top-up packs if you run over. The 1,500 from PH1000 are enough to run a sourcing-to-listing workflow and see where credits go.
SellerClaw
Spent most of this launch figuring out where to draw the line in the messaging between what SellerClaw does and what you still own. Each agent has a defined scope, irreversible actions require confirmed context, there's a full log. That's the architecture that earns the trust to expand the scope.