The Announcement
Cowboy, a New York startup building an app store for AI agents, emerged from stealth on November 4, 2026, with $5 million in seed funding. The round included Nebular, Compound, Semantic Ventures, Anagram, Shine Ventures, and Uncommon Projects, along with angel investors including Farnaz Azmoodeh, senior vice president of engineering at Function Health; Mats Olsen, co-founder and chief technology officer of Dune Analytics; Aashay Sanghvi; Kevin Kwok; James Smits; and Ian Lapham.
The company opened its marketplace, The General Store, the same day, pitching it as a place where developers can earn money from the agents and data feeds they build. The store launched with more than 100 agents and 10,000 data feeds for hire, with more than 10,000 agents, data feeds, skills, and models available across the platform.
The Founder
Cowboy was founded in 2025 by Chad DePue and venture fund Uncommon Projects, and is based in New York. DePue is chief executive officer and founder. He previously held engineering leadership roles at Whisper, Snap, and Uniswap — including scaling the engineering team at Uniswap, the decentralized crypto exchange — and he is credited with building Snap Kit, Games, Snap Minis, and Music during his time at Snap.
The company is a 10-person team incubated out of Uncommon Projects, a fund of former Snapchat managers including DePue. About two years after he left Uniswap, DePue built his own AI assistant to check email and run his calendar, and a former Uniswap colleague shared an agent he had built for crypto trading with DePue and close friends. That experience — a useful agent that could not easily be shared, sold, or trusted — became the seed of Cowboy.
What Cowboy Builds
Cowboy calls itself the app store for AI: a platform for developers to own, control, and monetize AI agents. Its central marketplace, the Cowboy General Store, provides access to autonomous agents, data feeds, models, computing resources, and reusable AI skills.
The key design choice is about trust. When someone hires an agent from the store, they get a private copy running in an isolated sandbox — a locally hosted version that does not share credentials or data back with the creator. Sellers get paid without giving up their source code or setting up payments. The platform includes a dedicated file system, secret store, and message queue, allowing agents to schedule work, hold assets, and communicate securely within isolated sandboxes.
Cowboy's first audience is traders, especially in crypto, who would pay to build or access agents that manage their positions. The store also opens with 10,000 data feeds for developers, plus a test version of Cowboy Chain, a blockchain-based payments network it plans to launch in the coming months. The company's mainnet — the live network layer meant to run the system — is planned for later in fall 2026.
The Problem Cowboy Is Solving
The AI agent economy has a trust problem. Agents need access to sensitive things — exchange accounts, email, calendars, payment credentials — to be useful. But buying an agent from a stranger means handing over the keys to your digital life, and selling an agent means handing over your code and hoping you get paid.
There is also an ownership problem. As DePue has put it, it remains unclear who owns agents, who secures them, and who gets paid when they work. Existing platforms are controlled by large companies that set the rules and take the cut. Cowboy's pitch is an alternative: a marketplace that big platforms do not control, where ownership, security, and payment are built into the infrastructure rather than left to trust.
The company frames this as an "App Store moment" for AI — the point at which a new computing platform gets its distribution layer, and developers can suddenly reach customers they could not reach before. The first app stores did not just distribute software; they created the economics of the mobile industry. Cowboy is betting that an agent store can do the same for the agent economy.
Why This Matters for Founders and Investors
The agent marketplace is becoming a crowded and contested space. Google brought partner-built agents into Gemini Enterprise's governed marketplace in April 2026, and AIR raised $50 million in September 2026 to vet agents and offer pre-screened components. At 100 agents and $5 million, Cowboy's launch is modest beside those efforts — and the company itself has acknowledged it will need much more funding than $5 million if the model works.
But Cowboy's angle is different from the enterprise marketplaces. It is building for individual developers and traders first, with privacy and ownership as the differentiators rather than enterprise governance. If the agent economy grows the way the app economy did, the distribution layer could be enormously valuable — and the company that builds it first, with the right trust model, could capture a lasting position.
The company's release does not state a valuation, and none has been disclosed. The platform's capabilities, security claims, and marketplace figures are self-reported, so investors should treat the 100-agent and 10,000-feed figures as launch-day claims rather than verified traction.
For founders building AI agents, Cowboy is worth watching as a potential distribution channel — and as a signal of where the agent economy's economics are heading. If developers can sell agents the way they sell apps, a whole new class of small software businesses becomes possible.
Angeled.com connects founders with investors.
Upstarts Media, November 4, 2026
Dealroom, November 4, 2026
Read Next
General Translation Raises $27 Million Series A Led by a16z to Build Agent-First Localization Infrastructure
LangChain Raises $300 Million Led by CapitalG at a $3 Billion Valuation to Build the Agent Stack
Mount Thor Raises $45 Million to Run Cloud Fleets of Macs for AI Agents
OpenAI Reportedly Raises $15 Billion at a $1.25 Trillion Valuation
Q-CTRL Raises $290 Million Series C Led by Salesforce Ventures, Reaching Unicorn Status
SE3 Raises $30 Million Series A Led by Accel



