Skip to content

Injection Surface Exposure

Metric: RCE & XSS Vulnerability Surface

Summary: A core Security Lens metric. It tracks external network I/O flowing near dynamic execution boundaries without adequate safety nets.

Effect: Illuminates the physical attack surface of the repository.

The Equation: Input vs. Execution

Step A: Define the Vectors $\(InputVectors = io\_hits + (ssr\_boundaries \times 2.0)\)$ $\(ExecutionVectors = (danger \times 4.0) + (safety\_neg \times 2.0)\)$

Step B: The Agentic RCE Spike (Prompt Injection) If an LLM orchestrator or Agentic Tool has direct proximity to an eval or OS command, it is an extreme vulnerability (Agentic RCE). * If Agentic logic + Danger logic are both present, \(ExecutionVectors\) are multiplied by \(10.0\), and the AI is classified as a hostile input vector itself.

Step C: Synthesize Mass $\(InjectionMass = (InputVectors \times ExecutionVectors) \times ArchetypeMultiplier\)$

Step D: Taint & Sigmoid Like the Logic Bomb, confirmed Tainted Injections add a massive +500.0 gravity spike before being routed through the standard Laplace-smoothed Sigmoid curve.



🌌 Powered by the blAST Engine

This documentation is part of the GitGalaxy Ecosystem, an AST-free, LLM-free heuristic knowledge graph engine.