EldorZ
Re:Is anyone else bothered that AI agents can basically do what they w...
on Monday April 27, 2026 @04:21AM (#47919029)
The control layer problem is real, and the enforcement
point matters. AegisProxy addresses runtime — what the
agent does after deployment. We've been working on the
pre-commit layer: what gets into the codebase before
an agent ever runs it.
The Replit incident is a good example — the agent wiped the database, but the conditions for that were set up earlier: no enforcement gate on what code and credentials could be committed. Sentinel Core sits at git commit and blocks before anything reaches the repository.
Complementary layers, not competing ones. https://datawizual.github.io
Show HN: Auditor Core–CLI security auditing engine with mathematical SPI scoring
https://github.com/auditor-core-systems/auditor-core-demo
I built a security auditing engine that combines 10 detection engines (Bandit, Semgrep, Gitleaks, IaC, CICD, dependency scanning and more) and produces a calibrated Security Posture Index instead of a raw findings dump.The scoring uses WSPM v2.2: SPI = 100 × e^-(Σ WeightedExposure / K) K scales dynamically with project size. Context matters — findings in test code are weighted differently...
Show HN: Auditor Core – Enterprise security auditing engine for DevSecOps
I built a deterministic security auditing engine that runs 10 detection engines (Bandit, Semgrep, Gitleaks, IaC, CI/CD, dependencies) and produces a calibrated Security Posture Index score. AI advisory via Gemini/Groq. Hardware-bound licensing. Would love feedback from security engineers.
Show HN: Sentinel Core – A zero-telemetry enforcement gate for GitHub Actions
Existing security scanners for CI/CD are mostly passive. They generate reports that developers often ignore. I built Sentinel Core to change that.It is a deterministic security engine that acts as a hard-fail gate. If a security invariant is violated, the build is blocked. Period.Key Features: Deterministic Enforcement: No more "warnings." It blocks unpinned Actions (CWE-1104), secret leaks,...
Show HN: Offline Deterministic Security Gate
Hi HN,I’m working on a security tool born out of frustration with how most security controls work today.In many environments, security happens after the fact: scan later, alert louder, hope someone reacts in time. Most tools also assume network access, dynamic updates, and trust in external services — which breaks down in regulated, air-gapped, or high-assurance environments.I decided to...