| Hey, I’m Mario. After chatting with a colleague about how AI agents are changing dev work, we got stuck on a question: Why share code when prompts can generate it on demand? I wanted to explore this further, so I build "Open Prompt Hub" — think GitHub, but for prompts: https://openprompthub.io How it Works: Instead of shipping binaries or source code, you share instructions and specs in form of a prompt. You can take this prompt, paste it into their agent or IDE and watch it build. If it’s not a perfect fit? Fork it, tweak it, and generate your custom version. All meta infos like version, description, test cases etc. are stored in a frontmatter block at the start of the prompt. So it's one file containing all the infos you need. (https://openprompthub.io/docs) Features of the platform:
- Versioned prompts with infos which models to best use
- Forking for customization
- Security Scans: prompts are scanned for security issues and prompt injections.
- User can give feedback, if the prompt successfully build what was promissed (scoped on models, so you know which one to best use for execution)
- Flagging mechanism It’s an MVP, but the core features—versioning, model-specific build status, and security scanning — are live. I'm currently looking into further features, such as:
- a git-like cli for publishing prompts and downloading/piping them directly to your agent
- multi stage/file prompts for more complex applications
- configurable prompts for e.g. switching programming languages, features, etc.
- better spec and test definition for build verification I’d love your feedback... on the idea, the spec and the platform. |