Appendix A: Glossary
Author: Shrikant Bhosale (@debtcollector21)
| Term | Definition |
|---|---|
| AttackGraph | Formal G=(V,E) model where V = security states, E = exploit steps. Enables pathfinding through vulnerability space. |
| Collapse | A finding is “collapsed” (rejected) when its total curvature κ_total < 0.50 or any dimension κ_d < 0.30. |
| Curvature (κ) | 6-dimensional geometric score of a finding’s reality. Product of AST, data flow, memory layout, error state, exploit chain, and temporal dimensions. |
| D_e | Information Debt of a function. D_e = 1 – H_e, where H_e is the fraction of error paths handled. |
| Debt Collector | The author handle (@debtcollector21) and the full 7-layer tool stack. |
| Error Principle | A system that does not handle its errors does not know its own state, and can be made to do anything. |
| Feynman Gate | 10-question honesty gate that every finding must pass before submission. Named after Richard Feynman’s “first principle” philosophy. |
| H_e | Error handling coverage. Fraction of error paths the programmer explicitly handled (0 to 1). |
| Information Debt | The entropy remaining after a programmer’s error handling logic. D_e = 1 – H_e. |
| KU | Knowledge Unit — a formal pattern description with pre/post conditions, CVSS, CWE mappings, and remediation. |
| NKU | Negative Knowledge Unit — created when a finding does not match any existing KU. Represents a gap in the pattern library. |
| Projection | The process of scoring a finding across 6 geometric dimensions. Each dimension tests a different aspect of exploitability. |
| Projection Collapse Theorem | A true vulnerability persists across ALL dimensions; a false positive collapses in one or more. |
| Quadratic Debt-Exploit Relation | P(exploit) ∝ D_e² because two unhandled errors must compose for exploitation. |
| UMM | Universal Memory Mapper — runtime + binary memory analysis detecting 8 debt patterns. |
| VMF | Vulnerability Manifold Framework — the static analysis engine. Treats vulnerabilities as points on a geometric manifold. |
| PAC | Pointer Authentication Code — ARM[Vendor].3 hardware feature that signs pointers before use. Shifts CFI debt from software to hardware key management. |
| MTE | Memory Tagging Extension — ARM[Vendor].5 hardware feature that tags memory granules. Probabilistic (4-bit = 15/16 detection). |
| SEP | Secure Enclave Processor — isolated coprocessor for cryptographic operations. Debt isolation strategy: narrows reachable surface. |
| Three-Gap Theorem | Theorem 0: D_IS > 0 AND D_SM > 0 AND D_MI > 0 for any non-trivial program. Intent, specification, and machine are never perfectly aligned. |
| Debt isolation | Strategy of walling off high-value state behind narrow interfaces (e.g., SEP). Reduces reachable D_e but concentrates debt at the interface boundary. |
| Debt hardening | Strategy of making exploitation more expensive (e.g., PAC, MTE). Reduces composition probability without changing source debt. |
| Toolchain-only vulnerability | Theorem 10: A vulnerability that exists only in the compiled binary, with no corresponding source expression. Created by LTO inlining + bounds check elimination. |
| JIT Double Debt | Theorem 8: D_JIT = 1 – (1 – D_spec)^N. JIT compilers accumulate debt at two layers (AOT + runtime speculation), creating TOCTOU windows. |
| D_e gate | CI/CD gate that blocks PRs or releases if D_e exceeds a threshold. The primary operational intervention for debt-managed engineering. |
| Debt budget | Quarterly allowance of D_e increase per team. Exceeding it triggers a debt-reduction sprint. Creates economic pressure to handle errors. |
| Debt-managed engineering | Engineering practice where Information Debt is tracked, trended, and gated like test coverage or memory usage. |