offensive-techniques/source-review-technique/SKILL.md
Hybrid AI/Deterministic SAST methodology for discovering zero-day vulnerabilities in source code. Orchestrates structural search with AI-driven data flow and sink validation.
npx skillsauth add aeondave/malskill source-review-techniqueInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Goal: Discover zero-day vulnerabilities in source code by combining deterministic structural search with LLM-assisted data flow analysis.
Modern SAST effectively bridges static tools with LLMs. The LLM acts as the triage and data-flow validator, while tools like semgrep or grep map the initial graph.
Do not ask the LLM to "find bugs" across thousands of lines at once. It will hallucinate.
exec, system, query, eval, deserialize, UnsafeCell, memory allocators.grep_search or vscode_listCodeUsages to find all invocations of the sinks.For each high-value sink found:
print is used", reject it unless you can prove the input is strictly user-controlled and unescaped.development
White-box auditing methodology for AI-generated ('vibe-coded') applications. Focuses on modern stack misconfigurations (Supabase, Next.js, Vercel).
development
Auth assessment: hardware/embedded methodology; UART/JTAG/SWD/SPI/I2C, firmware extraction, boot/debug paths, embedded OS evidence.
devops
Container methodology: Identifying containerization limits, Docker/K8s misconfigurations, and executing escapes to the host node.
development
CI/CD supply chain methodology: identifying poisoned pipelines, unsafe GitHub Actions, and extracting build secrets.