AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
MCP verified MIT Self-run

Processhacker Mcp

mcp-illegal-instruction-co-processhacker-mcp · by illegal-instruction-co

your ai debugger, vibe hacking tool

No reviews yet
0 installs
20 views
0.0% view→install

Install

$ agentstack add mcp-illegal-instruction-co-processhacker-mcp

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/mcp-illegal-instruction-co-processhacker-mcp)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
5mo ago

Declared compatibility

Claude CodeClaude DesktopCursorWindsurf

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Processhacker Mcp? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

processhacker mcp

this is dynamic mcp server for runtime analysis and process hacking. it is like processhacker but for ai agents.

setup

  1. open folder in visual studio (it has cmakelists)
  2. build all (ctrl+shift+b)
  3. it makes ProcessHackerMCP.exe and extensions/ folder.
  4. run exe. it communicates via stdin/stdout.

editor integration

you can configure your ai agent/editor to use this server. below are the mcp_config.json (or equivalent) settings. make sure to put the absolute path to the .exe.

cursor / gemini (antigravity) / claude desktop

add this to your mcp configuration file:

{
  "mcpServers": {
    "processhacker": {
      "command": "C:\\absolute\\path\\to\\ProcessHackerMCP.exe",
      "args": []
    }
  }
}

vscode (cline)

go to cline settings -> mcp servers and add:

{
  "mcpServers": {
    "processhacker": {
      "command": "C:\\absolute\\path\\to\\ProcessHackerMCP.exe",
      "args": []
    }
  }
}

> note: some editors might freeze if the mcp server sends a huge payload (e.g. reading 1GB of memory). the core now has a 2MB payload protection limit, but try to use limit and offset arguments when querying big processes.

agent guardrails & telemetry (new in v1.1.0)

we added some enterprise-grade agent logic to stop rogue bots from nuking your host machine.

  • Read-Only Mode: change "args": [] to "args": ["--read-only"] in your mcp config. if the ai tries to write memory or suspend threads (destructive actions), the core blocks it.
  • Audit Log: all tool calls (and their args) are saved to processhacker_audit.log. destructive actions are tagged with [WARNING: DESTRUCTIVE].
  • Loop Breaker (Rate Limit): if an ai agent panics and calls 50 tools in 1 minute (brute-forcing memory), the core locks it out for 30 seconds. write a c++ extension for heavy scanning, don't spam rpc.
  • Loud Failures: if reading unmapped memory fails, the ai gets a clear hint to use ph_query_memory_regions instead of just a generic error.

how to make extension (for bypass etc)

core is just router. all tools are in dll plugins.

if u want make stealth bypass (like vehbutnot or direct syscall):

  1. copy extensions/sample_ext folder.
  2. write your code in c or c++.
  3. setup McpToolRegistration and set isDestructive = true if your tool mutates state (writes memory, sets hooks).
  4. u must export InitMcpExtension.
  5. put your compiled .dll inside extensions/ folder.
  6. exe will auto load your dll on start and ai agent will see your new tool.

> new in v1.6.0: the ai agent can now write its own extensions dynamically! by using the ext_auto_compiler tool, the agent can send raw C code which the router compiles using a bundled TCC (Tiny C Compiler) and hot-loads directly into active memory. you can literally ask the ai to write its own custom bypass and it will compile itself on the fly!

contribute

if u write good stealth extension and think it can bypass anything or help others, please send pull request (pr). we need more plugins for stealth.

just wondering... (faq)

i was thinking about this architecture and had a weird thought:

could someone actually use this to create autonomous malware or game cheats just by writing prompts? like asking the ai: "inject here, find the decryption routine, and dump the keys as json". since the actual "malware behavior" isn't in the compiled c++ code but in the prompt text, no classic anti-virus could catch the payload statically.

and if an anti-cheat updates, the ai could just read the new memory layout and adapt its logic instantly without needing a recompile.

is this genuinely possible now or just a weird architectural nightmare? lol if u have thoughts on this, hit me up or open an issue.

disclaimer

educational and research purposes only. published to document the technique, not to hand it out as a toolkit.

what you do with this is your problem. no warranty, no support, no liability.

license

MIT. do whatever. don't blame us.

Source & license

This open-source MCP server is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.