Updated November 18, 2026
Glossary ToolsBest models

Task · CTF

AI for CTF reversing challenges

For CTF reversing, use an LLM to read the check, explain the transform and write a solver. Check the answer by running it in the binary.

By AI for Reverse EngineeringUpdated November 18, 2026
On this page

Rules

Read the event's rules on AI assistance before you start.

Workflow

  1. Find the check from the success and failure strings and their cross-references.
  2. Decompile the check and its callees, and rename what you already understand.
  3. Ask what happens to each input byte.
  4. Ask for the inverse as Python you can read.
  5. Run it. The binary is the judge, not the model.

Prompt

Adapted from the ida-pro-mcp README, for any MCP server.

Prompt

Analyze this crackme with the MCP tools. Comment and rename as you go. Use int_convert for number bases. Do not brute force. Write report.md and the password.

ida-pro-mcp README

Tools

ToolWhy it suits CTFs
GhidraMCPFree with Ghidra. The agent decompiles and renames
IDA MCP / ida-pro-mcpThe prompt above comes from its README
Binary Ninja MCPIts README links a CTF demo
r2ai and r2copilotr2ai in the shell, r2copilot for CTFs

Pitfalls

  • A 32-bit little-endian constant read backwards breaks the solver.
  • A right shift changes meaning when the value is signed.
  • Check loop bounds in the disassembly, and use the tool's number converter.

On a pwn challenge, a model can read the bug and the protections. This site does not cover writing the exploit.

Write-up

After the event, hand the model your notes. With 943,718 output tokens, Kimi K3.1 Pro can draft the write-up in one reply.

Scope

Authorized work only, on your own systems, a scoped engagement, a CTF or a lab. No exploits, payloads, malware or detection-evasion steps.