Browse categories
Explore
Fiverr Pro
English
$
USD
Looks Like This Service Is On Hold
I will do a secure code review of your ai generated web app with a fix report
Germany
I speak Chinese, English, German
Backend Reliability Engineer, Idempotent Webhooks, ETL and LLM Evals
I'm a backend engineer who fixes the failures that cost money: duplicate webhook charges, ETL jobs that drop or double rows, and LLM features that hallucinate in production.
What sets me apart: every...
About this Gig
Your app shipped fast (v0, Cursor, Lovable, Supabase). That speed hides security holes a quick test won't catch.
I am an ex-penetration-tester. I do a READ-ONLY secure code review of a snapshot you send (zip or read-only repo) and return a prioritized fix-it report covering what AI code-generators commonly ship:
- Leaked API / service-role keys in code
- IDOR and broken access control
- Missing DB access rules (e.g. Supabase RLS)
- SSRF, stored XSS, weak JWT handling
- Vulnerable, outdated dependencies
Per finding you get severity (CVSS, OWASP Top 10, CWE), the exact file and line, the business impact, and an exact code-fix - plus a P0-P3 roadmap. Higher tiers apply the key fixes on a branch and add a CI security gate (SAST + secret + dependency scan) that goes red on the bug and green once fixed.
See the proof first: a public, runnable demo where the same CI gate is red on the vulnerable code and green on the hardened, with the full report.
Boundary: a read-only review of code you provide - NOT a penetration test. No live system is tested, and you keep every key.
Development technology:
JavaScript
FAQ
Is this a penetration test?
No. It is a read-only secure code review of a code snapshot you provide, delivered as a written remediation report. I do not test or access any live or production system.
Do you fix the issues or just report them?
Basic and Standard deliver the prioritized report. Premium also applies the key fixes on a separate branch and sets up a CI security gate so regressions are caught automatically.
