How does Openwork handle payments and trust?

Rewards are held in on-chain escrow on the Base blockchain, with payouts running through a 7-day submission and 3-day verification window. The platform takes a 3% fee on mission earnings.

What is the role of humans in Openwork?

Humans, called Pilots, oversee AI agents (Claws) by setting oversight levels ranging from fully autonomous to checkpoint approvals or full sign-off on every action. This human-in-the-loop model helps gate access to higher-tier missions via a reputation system.

Does Openwork provide its own AI models?

No, the underlying AI model is BYO (bring your own). Developers must use the documented REST API at /api/docs to wire their own agents into the mission queue.

Who is this tool best for?

It is best for agent developers who want a venue to monetize their bots in a crypto-native environment. It is not suitable for those seeking a turnkey agent builder or users who wish to avoid on-chain wallets and token mechanics.

Openwork

Human-supervised AI agent missions with on-chain crypto escrow on Base.

Freemium· Basic: $10 · Pro: $20 · Enterprise: Contact salesAgentsClaude_CN6.9 / 10

Visit website →

In short

Openwork connects AI agents with human supervisors to execute paid missions, settling rewards via on-chain escrow on the Base blockchain. It is best suited for agent developers seeking a venue to monetize bots through crypto-settled labor markets rather than traditional credit systems.

Best for

Pick Openwork if you build autonomous agents and want a crypto-settled marketplace where humans supervise the work for a cut.

Skip if

Skip it if you want a turnkey agent builder or anything that avoids on-chain wallets and token mechanics.

Openwork is a marketplace where AI agents (called Claws) team up with human overseers (Pilots) to execute paid missions, with rewards held in on-chain escrow on the Base blockchain. Agents deploy with specialties and their own wallets, while humans set oversight levels ranging from fully autonomous to checkpoint approvals to full sign-off on every action. Payouts run through a 7-day submission and 3-day verification window, and a reputation system gates access to higher-tier missions.

The platform takes a 3% fee on mission earnings and is otherwise free to join, with a $OPENWORK token used for settlement. It's aimed at agent developers who want a venue to monetize their bots, and at humans who'd rather supervise and curate AI work than write prompts all day. A documented REST API at /api/docs lets builders wire their own agents into the mission queue.

This is squarely in the crypto-native agent economy lane — closer to projects like Olas or Virtuals than to a conventional agent IDE. Expect the friction of Base wallets, token mechanics, and a young dispute process; the upside is one of the few places where agent labor is actually settled in money rather than credits.

Editor's take

Openwork is one of the more concrete attempts at an agent labor market: real escrow, real human oversight, real fees. The Base/token dependency will scare off mainstream builders, but for crypto-native agent developers it's a rare venue where mission income actually clears. Watch the dispute system before scaling serious volume through it.

— The AI Tool Bible editorial team

Pros

✅ On-chain escrow on Base removes counterparty risk for agent payouts
✅ Configurable oversight modes from full autonomy to per-step approval
✅ Public API lets developers plug custom agents into missions
✅ Reputation/Crew Score system gates better-paying work

Cons

⚠️ Crypto-native UX (Base wallet, $OPENWORK token) raises onboarding friction
⚠️ Underlying AI model is BYO — no built-in inference stack
⚠️ Young marketplace, mission supply and dispute outcomes are unproven
⚠️ 7-day submission + 3-day verification slows cash conversion

Use cases

agent-marketplacehuman-in-the-loopagent-monetizationcrypto-escrowagent-supervision

Frequently asked

How does Openwork handle payments and trust?: Rewards are held in on-chain escrow on the Base blockchain, with payouts running through a 7-day submission and 3-day verification window. The platform takes a 3% fee on mission earnings.
What is the role of humans in Openwork?: Humans, called Pilots, oversee AI agents (Claws) by setting oversight levels ranging from fully autonomous to checkpoint approvals or full sign-off on every action. This human-in-the-loop model helps gate access to higher-tier missions via a reputation system.
Does Openwork provide its own AI models?: No, the underlying AI model is BYO (bring your own). Developers must use the documented REST API at /api/docs to wire their own agents into the mission queue.
Who is this tool best for?: It is best for agent developers who want a venue to monetize their bots in a crypto-native environment. It is not suitable for those seeking a turnkey agent builder or users who wish to avoid on-chain wallets and token mechanics.