Individual skill

web-python-sandbox-escape

Escape a Python sandbox / eval jail to reach code execution — defeat keyword blocklists and restricted eval/exec by reaching objects through the class hierarchy. Load when user input hits eval/exec/a "safe" expression evaluator, a Python REPL/calculator feature, a template that runs Python, or a filtered code box. Signals: "eval", "exec", `__import__` blocked, banned words (import/os/system), a Python jail, PyYAML/pickle input, a formula/expression field.

skills/web/web-python-sandbox-escape/SKILL.md

Read pinned source

Source commit ee0e0c50d2f6d941a14b73dd85a2b470d0ab976e

What was checked

Matches the observed repository HEAD: Yes
Scanner rules are current: No
Repository scan complete: No
Skill scan complete: Yes
Observed behavior: Not checked yet
Last checked: Nov 7, 2026

These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.

Install this skill

Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.

An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.

Repository context

Repository stars (not skill ratings): 27
Repository last updated: Oct 14, 2026
License: MIT

Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.

agent-skills Repository topic
ai-agents Repository topic
ai-security Repository topic
application-security Repository topic
blue-team Repository topic
bug-bounty Repository topic
claude-code Repository topic
codex Repository topic
detection-engineering Repository topic
gemini Repository topic
engineering Derived category

Files and excerpt

File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.

Browse files

SKILL.mdcore · 3,416 bytes

Read excerpt


# Python sandbox / eval escape

## When it applies
User input reaches `eval()`, `exec()`, a "safe expression" evaluator, a template engine that runs
Python, or a feature that runs user formulas — and a blocklist tries to make it safe by banning
words like `import`, `os`, or `system`. If you can evaluate Python at all, the blocklist is almost
always bypassable.

## Why it works
String-based keyword filters remove *names*, but the objects they'd name are still reachable through
Python's object graph. From any object you can walk `__class__ → __base__ → __subclasses__()` to
find a class whose module gives you code execution (`subprocess.Popen`, `os.system`, `builtins`),
without ever typing a banned identifier literally.