TimeCodeSecurity – Deterministic AST SAST and Auto-Remediation for Python

Developer Ayush Gaur released TimeCodeSecurity (TCS), an open-source static application security testing (SAST) engine for Python that uses deterministic Abstract Syntax Tree (AST) taint tracking instead of regex patterns or LLMs. TCS tracks dataflow from function parameters into sensitive execution sinks including subprocess.run, raw SQL cursors, and path operations, and validates remediation patches via AST syntax compilation and in-memory re-scanning before writing to disk, running in roughly 700ms per scan. The tool is installable via pip from its GitHub repository and was posted to Hacker News, where it received 2 points and 0 comments.

Hi Hacker News I built TimeCodeSecurity TCS , an open-source static security engine for Python. Most existing open-source SAST tools rely on regex patterns or LLMs, leading to high false-positive rates and alert fatigue. TCS takes a compiler-level deterministic approach: it parses Python's Abstract Syntax Tree AST and tracks dataflow from function parameters directly into sensitive execution sinks subprocess.run, raw SQL cursors, and path operations . Key highlights: - Deterministic AST taint tracking Zero regex / Zero AI guessing - Closed-loop remediation: Patches are validated via AST syntax compilation and re-scanned in-memory before touching disk - Chronological visual proof graphs: Source - Taint - Sink - Sub-second execution ~700ms You can run it directly in your terminal: pip install git+https://github.com/kushigaur3103-svg/time-code-security.git tcs scan ./your project --fix GitHub: https://github.com/kushigaur3103-svg/time-code-security Would love feedback on the AST traversal and in-memory verification approach https://github.com/kushigaur3103-svg/time-code-security And that's my linkdin link https://www.linkedin.com/in/ayush-gaur-1488b1420?utm source=share via&utm content=profile&utm medium=member android Comments URL: https://news.ycombinator.com/item?id=49765426 https://news.ycombinator.com/item?id=49765426 Points: 2 Comments: 0