A new self-hosted application security platform called Eatheria has emerged on Hacker News, promising to tackle one of the most persistent pain points in DevSecOps workflows: the avalanche of false positives that buries real vulnerabilities. The project, shared via a Show HN post by developer MartinCrespoC, positions itself as an alternative for teams that want full control over their security scanning infrastructure rather than relying on cloud-based SaaS offerings.

What Eatheria Brings to the Table

The platform's core differentiator is its AI-powered false-positive filtering layer. Traditional SAST and DAST scanners tend to flag everything that looks remotely suspicious, forcing security teams to wade through hundreds or thousands of alerts to find genuine issues. Eatheria apparently uses machine learning to analyze scan results and suppress noise before they ever reach a human reviewer. The self-hosted angle means organizations keep their code and vulnerability data within their own environment—no third-party cloud processing required.

Why Self-Hosted AppSec Matters Right Now

Enterprise teams have increasingly gravitated toward self-hosted solutions as supply chain concerns and data sovereignty requirements tighten. Sending source code or binary artifacts to external scanning services creates compliance headaches for regulated industries, particularly in finance and healthcare. A platform that runs on-premises gives security engineers the ability to audit exactly what their tools are doing with sensitive intellectual property—something that's harder to guarantee with vendor-managed pipelines. The AI filtering component addresses a practical bottleneck: developer trust erodes when scanners cry wolf too often. When teams stop paying attention because everything is marked critical, actual vulnerabilities slip through. Reducing that noise without suppressing true positives requires careful ML model design, and how well Eatheria executes this will likely determine whether it gains traction beyond early adopters.

Getting Started

The project repository is available at the GitHub link in the source below. Given the Show HN context, expect a project that's functional but potentially rough around edges—typical for tools coming from independent developers rather than funded startups.

Key Takeaways

  • Eatheria targets self-hosted AppSec workflows with an AI layer to reduce false positives
  • Appeals to teams with strict data residency or supply chain security requirements
  • Early-stage project posted on Hacker News by individual developer MartinCrespoC
  • Success depends heavily on the accuracy of its ML-based filtering approach

The Bottom Line

The AppSec space desperately needs tools that respect developers' time, and false-positive fatigue is a solved problem waiting for a widely-adopted solution. Eatheria's self-hosted model is smart positioning—if the AI filtering actually works well in practice, this could be worth watching as it matures.