A Claude Code plugin for finding semantic security flaws in CPython's standard library — archive extraction boundaries, validation coverage gaps, resource amplification, and audit hook bypass.
This repository also configures its own agents. See what cpython-security-toolkit tells them →
ReviewToolkits/cpython-security-toolkit
Agent Claude Code
Detects file-loading paths that bypass sys.audit() hooks by using open() instead of io.opencode(), and shell-calling paths where validation occurs before template substitution. Invoke when scanning importlib, webbrowser, or venv.
ReviewToolkits/cpython-security-toolkit
Agent Claude Code
Detects decompression paths that materialize unbounded output before applying a size check, and allocations whose size is directly controlled by attacker-supplied archive metadata. Invoke when scanning zipfile, tarfile, lzma, bz2, gzip, or plistlib.
ReviewToolkits/cpython-security-toolkit
Agent Claude Code
Analyzes CPython security fix commits and checks whether the fix was applied to all code paths through the same invariant. The most important agent for preventing second-round CVEs. Invoke after identifying a security fix commit or when reviewing recent type-security labeled changes.
ReviewToolkits/cpython-security-toolkit
Agent Claude Code
Detects archive extraction paths that may write outside the destination directory. Invoke when scanning tarfile, zipfile, or shutil.unpackarchive for path traversal vulnerabilities.
ReviewToolkits/cpython-security-toolkit
Agent Claude Code
Detects security-sensitive value types where some assignment paths validate input but others do not. Specializes in the incomplete-fix pattern. Invoke when scanning http.cookies, wsgiref, urllib, or webbrowser for validation gaps.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: