FablePool
Pool money behind a prompt and an AI agent attempts the build in public
Visit website →About
FablePool is a public-build platform where supporters pool funds behind a single prompt, and an AI agent attempts the project step by step in the open. Each milestone fills, runs, and ships whatever it produced — no closed-door delivery.
Every project keeps a double-entry ledger of money in, tokens spent, and artifacts out. Entries are append-only, so corrections show up as new lines instead of edits, and anyone can audit how a build actually used its budget.
Builds run on Anthropic's Claude Fable 5, and the output ships open-source under the MIT license. Pricing is credit-based: 1 credit equals one cent of inference, applied to a project's pool until a milestone funds and the agent runs.
AI-assisted draft, human-reviewed before publishing — see how we choose & review tools.
Why we picked it
This is a curiosity pick, not a work tool — FablePool is worth a look for the transparency mechanic: a public double-entry ledger of money and tokens spent as an agent attempts a funded build in the open, rather than any promise about what the agent can actually ship.
Related tools
- 1 withHuman Human approval layer for AI agents built on an open protocol Freemium
- 2 guilds.run Open-source AI agent teams you can actually inspect Free
- 3 OpenSwarm An AI-first OS where agent swarms do the work in parallel Free
- 4 Opengeni Open-source infrastructure for AI agents in your product Freemium
- 5 iFixAi Independent auditing for AI agents that catches hidden misalignment Freemium
- 6 Recurse Serverless harness that lets your agents compose specialists Freemium