CTF — Binary Exploitation (Pwn)

What it does

This skill equips an agent with an extensive quick-reference and playbook for CTF-style binary exploitation (pwn) challenges. It consolidates concise one-line techniques and detailed supporting notes covering stack and heap overflows, format-string attacks, ROP chain construction, advanced allocator primitives, kernel exploitation patterns, sandbox/VM escapes, and practical debug/tooling commands. The content is oriented at hands-on exploit development and triage during CTFs or vulnerability research.

When to use it

Invoke this skill when faced with a memory-corruption or low-level vulnerability: untrusted binaries, native services, kernel modules, or constrained sandboxes. Use it during live CTF rounds, writeups, or when the agent needs concrete exploit primitives (find gadgets, build ROP, leak libc, or craft shellcode). Pivot to related skills (reverse, web, crypto, misc) as the investigation indicates.

What's included

• Scripts: none bundled, but the skill references many markdown playbooks and tooling commands (pwntools, ROPgadget, ropper).
• References: extensive markdown reference files (overflow-basics, rop-and-shellcode, heap-techniques, kernel, sandbox-escape, etc.).
• Instructions: quick start commands for analysis and debugging, decision trees for protection strategies, and a catalog of exploit primitives and mitigation implications.

Compatible agents

Best used with filesystem-capable agents that support shell and Python (e.g., Claude Code, similar agents with bash/python access).