Skip to content

Logic Bomb Exposure

Metric: Sabotage Mass (Delayed Execution & Destruction)

Summary: A core Security Lens metric. Logic bombs sit dormant until a specific condition is met, at which point they execute a destructive payload. This metric hunts for condition-heavy code that ends in system halts, bailouts, or aggressive execution.

Effect: Highlights files containing highly conditional destructive commands.

The Equation: Trigger & Payload

We calculate Sabotage Mass by multiplying the "Trigger" (the conditional logic) by the "Payload" (the destructive outcome).

Step A: Define the Trigger We weigh standard branching heavily against thread-halting commands (which are often used to delay execution). $\(Trigger = branch\_hits + (halt\_hits \times 3.0)\)$

Step B: Define the Payload We look for system exits, manual memory cleanup, and dynamic execution. $\(Payload = (bailout \times 2.0) + (cleanup \times 1.5) + (danger \times 4.0)\)$

Step C: The Agentic & Hardware Shields AI orchestration and hardware bridges naturally use dynamic execution and halting. We divide the Payload by the presence of these heuristics to prevent false positives in robotics/ML repositories.

Step D: The Taint Spike If the static engine explicitly confirmed that tainted input flowed directly into an execution sink, this is an absolute vulnerability. $\(SabotageMass += (TaintedInjection \times 500.0)\)$

Step E: Sigmoid Mapping The \(SabotageMass\) is normalized against the padded LOC and mapped to the 0-100 scale using the security sigmoid curve.



🌌 Powered by the blAST Engine

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