IDA Pro MCP with a model that doesn’t refuse

ida-pro-mcp bridges IDA Pro to any MCP client, so a coding agent can decompile functions, read disassembly and rename symbols. Mainstream models often refuse to reverse a binary. Pair ida-pro-mcp with a coding agent running an Obscurify.ai model, and the work gets done. The account is a 16-digit number, not an email.

Authorized work only. Reverse engineer your own binaries, CTF challenges, malware you are analyzing, and software you are licensed or legally entitled to examine. Obscurify.ai’s terms prohibit breaking into systems you aren’t authorized to access. You also need a legitimate IDA Pro license; Obscurify.ai grants no tool license.

The agent-plus-model half is tested: on 2026-11-08, OpenCode running the pqhaz/deepseek-v4.1-flash-1017-heretic model through Obscurify.ai reverse engineered a crackme we compiled ourselves and recovered its hardcoded password. The ida-pro-mcp wiring below is verified against the project’s README on 2026-11-08 and was not run here (IDA Pro is a commercial tool not installed on our test machine).

Before you start

  • IDA Pro (a licensed install) and Python.
  • ida-pro-mcp: pip install git+https://github.com/mrexodia/ida-pro-mcp, then ida-pro-mcp --install to add the IDA plugin. Start the MCP server from inside IDA (Edit → Plugins → MCP) or via the packaged command.
  • A coding agent that speaks MCP, such as Cline or OpenCode, already pointed at Obscurify.ai.
  • An Obscurify.ai API key. Create an account — a 16-digit number, no email. Add credit, then Show API Key.

Add the MCP server to your agent

ida-pro-mcp exposes a local MCP endpoint. Register it in your agent’s MCP config, for example:

{
  "mcpServers": {
    "ida": {
      "command": "uvx",
      "args": ["ida-pro-mcp"]
    }
  }
}

Check ida-pro-mcp’s README for the current command and port; it also prints a ready-made client config on install.

Pick a model

A model that won’t refuse and supports tools. From /v1/models:

  • Uncensored, needs the Pro plan: deepseek/deepseek-v4.1-flash-uncensored or z-ai/glm-5.3-flash-uncensored.
  • Decensored / cybersecurity, any funded account: pqhaz/deepseek-v4.1-flash-1017-heretic, orca/orcacyber-minimax-m3.1-flash, heabsy/cyberheabsy.

Use it

Load your target in IDA, then ask your agent: “Decompile the function at the entry point, follow the authentication check, and explain it. This is my own binary.” The agent calls IDA through MCP and reasons over the results with your Obscurify.ai model.

If something fails

  • 402 / 403: no balance, or the model isn’t on your plan (uncensored needs Pro).
  • Agent can’t see IDA tools: confirm the MCP plugin is loaded in IDA and the server is running.
  • 404 model_not_found: check the id against /v1/models and keep its prefix.

What stays private

Prompts and responses aren’t stored on Obscurify.ai’s servers; models with the ZDR tag run only on zero-data-retention routes. Providers see the prompt text — including decompiled code your agent sends — but not who sent it. See the privacy policy.

More: Obscurify.ai for agents.