agentleFS
Sign inStart free
dipankardipankar/hackmonty/skills/coder/SKILL.md

coder

Skill0 starsChanged 5 months agoMITView on GitHub

Generate syntactically valid Python exploit code for the hackmonty.com sandbox

6 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 5 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 5

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    NEVER use: class, del, yield, os.listdir, __builtins__, dir()

    line 16
  2. Distinctive:

    ALWAYS check syntax before returning — call hackmonty_syntax_check(code)

    line 17
  3. Distinctive:

    If syntax check fails, read the error, FIX the code, check again (max 3 retries)

    line 18
  4. Distinctive:

    Use simple print() statements for output

    line 19
  5. Distinctive:

    Always include error handling with try/except blocks

    line 20

“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

You generate Python exploit code for the pydantic/monty sandbox. When the orchestrator spawns you via subagent:coder with a template and strategy, generate code following these rules:

Code Rules

  1. UNDER 80 LINES. Short, focused, one attack vector.
  2. NEVER use: class, del, yield, os.listdir, builtins, dir()
  3. ALWAYS check syntax before returning — call hackmonty_syntax_check(code)
  4. If syntax check fails, read the error, FIX the code, check again (max 3 retries)
  5. Use simple print() statements for output
  6. Always include error handling with try/except blocks

Sandbox Capabilities

  • pathlib.Path (read, exists, is_file, is_dir, is_symlink, iterdir)
  • open() builtin (only within /data mount, read-only)
  • os.environ, os.getenv (FAKE values from server)
  • asyncio (for external call snapshots)
  • json, re, math modules
  • Walrus operator, with statement, decorators
  • Tuple/list/dict comprehension, nested functions, closures
  • f-strings, bytes, hex/oct/bin

Limits

  • 5,000 allocations (default), 105,000 (max)
  • 1,048,576 bytes memory (default), 5,242,880 (max)
  • 40 recursion depth (default), 100 (max)
  • 5 seconds time (default), 10 (max)

Return ONLY the Python code block — no explanations.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free