Sentrint, a security scanner for projects build with LLMs
Scans repos built with AI coding tools for exposed secrets and security holes.
Sentrint points its scanner at the repo behind an app built with Claude, Gemini or any of 16 LLM platforms, checking for hardcoded secrets, open database rules, vulnerable dependencies and exploitable code paths. An AI layer drops findings that aren't actually exploitable and writes the fix as a prompt for whichever platform built the app — paste it back in, rescan, and the score climbs. It runs on a single-use Google Cloud instance with read-only repo access, deleted after each scan.
What holds up
- +Free first scan — paste a GitHub repo URL, no signup barrier mentioned.
- +Writes the fix as a ready-to-paste prompt for the exact tool that built the app.
- +Clones the repo to a single-use instance with read-only access, then deletes it.
Mind the limits
- −Doesn't fix anything itself — you paste its fix prompt back into your tool by hand.
- −Requires giving a cloud-hosted scanner read access to your repo, even briefly.
Featured here? Take the badge
Put it on your site — it links back to this review. Free for every listed product, always.
<a href="https://stillworks.dev/products/p/sentrint-a-security-scanner-for-projects-build-with-llms/"><img src="https://stillworks.dev/badge/sentrint-a-security-scanner-for-projects-build-with-llms.svg" alt="Picked by StillWorks" width="250" height="54"></a>