Hi Hacker News! Matvey, one of the authors, is here. While building enterprise agents, we ran into a problem: the more tools you connect to the AI, the higher the chance it will run out of control and leak sensitive data. Guardrails, in theory, should prevent this, but the situation is worrying: - Non-deterministic guardrails (LLM as a judge, auto modes, etc.) are vulnerable to prompt injections, or they lack knowledge of the data, making them inefficient (~10% data leaks on our benchmarks). - Existing deterministic guardrails (Cedar, OPA, FIDES, Dogwood) require massive case-specific IF-E
Thinking of building something like this?
Every launch here is a competitor to somebody's idea. If yours is close, check it against the market before you build: the Full Check names the rivals, the prices and the gaps.
More developer tool / api launches
All
ElevenLabs UI
Audio and agent components for Next.js built on shadcn/UI
What the Font
Identify and discover fonts from images.
Wispbit
Linter that enforces codebase standards with AI coding agents.
OnlyJPG
Private browser-based converter for any image format to JPG.
Scriber Pro
Offline AI transcription app for macOS with no cloud uploads.
Duck-UI
Browser-based SQL IDE for DuckDB running entirely in WebAssembly.
Checked ideas in SaaS & software
AI phone receptionist for small clinics in Canada Kill
A voice AI that answers calls, books appointments and sends reminders for small Canadian physio and dental clinics at C$149 a month.
Browser extension that summarises Terms of Service Kill
Free Chrome extension that turns any site's terms and privacy policy into five plain bullets, with a $4 a month pro plan.
AI bookkeeping assistant for freelance designers Kill
A $19/month app that links a designer's bank and invoicing tools, sorts expenses and prepares quarterly tax estimates.