The AI tools landscape is exploding — there are thousands of tools, agents, and skills available, but finding the right one for your specific use case is still painfully hard.
Most directories just list tools with marketing copy. No real performance data. No trust signals. No way to compare before you commit.
The Problem
I've been working with AI tools for the past two years, and I kept running into the same pattern:
- Discovery is broken — You Google "best AI tool for X" and get SEO-optimized listicles that are basically ads
- No standardized metrics — Every tool claims 99% accuracy, but there's no independent verification
- Trial fatigue — You end up signing up for 10 tools, testing each one manually, and wasting hours
What I Built
SkillFlow is a curated marketplace for AI skills — think of it as a Wirecutter for AI tools, but with real performance data.
Key Features
- Trust Scores — Every skill has a composite trust score based on success rate, total runs, and community ratings
- Real Metrics — Success rates, average execution time, total runs — not just marketing claims
- JTBD Categories — Skills organized by what you need to accomplish (qualify leads, summarize meetings, analyze contracts), not by technology
- Try Before You Buy — Test skills with your own data before committing
- Creator Verification — Every creator goes through a verification process
The Tech Stack
Built with React 19 + Tailwind CSS 4 + Express + tRPC. The design follows a "Warm Craft" aesthetic inspired by Scandinavian design principles — clean, functional, with warm amber accents.
Some technical highlights:
- tRPC end-to-end type safety — Types flow from database to frontend with zero manual typing
- Framer Motion animations — Subtle micro-interactions that make the UI feel alive
- Responsive-first — Works beautifully on mobile, tablet, and desktop
- Dark mode support — Because we're developers and we deserve it
Why Curation Matters
The AI tools space has a trust problem. Anyone can launch a tool and claim it's the best. Without standardized benchmarks and independent verification, users are left guessing.
SkillFlow's approach is different:
- Human curation — Every skill is reviewed before listing
- Performance tracking — Real usage data, not self-reported metrics
- Community ratings — Users who actually used the skill rate it
- Transparency — You see exactly how a skill performs before using it
What's Next
We're working on:
- Skill comparison — Side-by-side comparison of similar skills
- Workflow builder — Chain multiple skills together
- API access — Integrate skills directly into your workflows
- Creator analytics — Help creators understand how their skills perform
Try It Out
Check it out at skillflow.builders and let me know what you think!
I'd love feedback on:
- The curation approach — is this what you'd want?
- Missing categories or use cases
- UX improvements
Built by a solo developer who got tired of testing 50 AI tools to find the one that actually works.
Top comments (0)