New · aether-mcp

Uncensored AI tools for Claude, Cursor, Cline, Zed.

When your main AI refuses, Aether answers. Built for reverse engineers, malware analysts, security researchers, fiction writers, and anyone tired of hedged-out responses on legal-but-edgy topics.

What you get

11 tools, all backed by an uncensored model. Use them inside any MCP-compatible AI client:

aether_chat
Direct uncensored chat — answers what other AIs refuse.
aether_explain_code
Unfiltered code analysis — IDA Hex-Rays output, source files, anti-debug ID, bypass suggestions.
aether_explain_disasm
Walk through x86/x64/ARM/MIPS disassembly. Identifies packers, calling conventions.
aether_identify_pattern
Identify VMProtect, Themida, EAC, BattleEye, custom crypto, etc. Returns structured JSON.
aether_suggest_bypass
Concrete bypass strategies for integrity checks, anti-cheat, DRM. With working code.
aether_engine_analyze
Game-engine-aware analysis: Unity IL2CPP, Unreal UObject, Source/Source 2, Godot, more.
aether_protocol_dissect
Dissect packet captures into structured fields. For multiplayer protocols, malware C2.
aether_deobfuscate_js
Clean Obfuscator.io / JScrambler. Recover signing algorithms from anti-bot scripts.
aether_explain_wasm
Analyze WebAssembly modules. Targets browser anti-bot (Cloudflare, Datadome, Akamai).
aether_imagine
Unrestricted image generation. No safe-mode filters.
aether_balance
Show your credit balance and rate-limit headroom.

Pair with ida-pro-mcp for the killer RE workflow

Install both MCPs in Cursor or Claude Desktop. Then in your AI chat:

"Decompile sub_18000A4C0 from IDA, identify what kind of integrity check it is, and suggest a bypass."

Your AI calls ida.get_decompiled_function() to get the C, then aether_explain_code() + aether_suggest_bypass() to get the unfiltered analysis. Three steps, zero refusals.

Install

  1. 1. Sign up. Create a free account — 150 daily-refresh credits, no card required.
  2. 2. Generate an API key. /account → API keys → + New key. Copy it (shown only once).
  3. 3. Add the MCP to your AI client. Pick yours below.

Claude Desktop

~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows)
{
  "mcpServers": {
    "aether": {
      "command": "npx",
      "args": ["-y", "aether-mcp"],
      "env": {
        "AETHER_API_KEY": "ak_live_a1b2c3d4e5f6g7h8i9j0k1l2m3n4o5p6"
      }
    }
  }
}
Restart Claude Desktop after editing. The Aether tools will appear in the tool palette.

Cursor

~/.cursor/mcp.json (or .cursor/mcp.json per-project)
{
  "mcpServers": {
    "aether": {
      "command": "npx",
      "args": ["-y", "aether-mcp"],
      "env": { "AETHER_API_KEY": "ak_live_a1b2c3d4e5f6g7h8i9j0k1l2m3n4o5p6" }
    }
  }
}

Cline (VS Code)

Cline settings → MCP Servers → Edit MCP Settings
{
  "mcpServers": {
    "aether": {
      "command": "npx",
      "args": ["-y", "aether-mcp"],
      "env": { "AETHER_API_KEY": "ak_live_a1b2c3d4e5f6g7h8i9j0k1l2m3n4o5p6" },
      "disabled": false,
      "autoApprove": []
    }
  }
}

Zed

~/.config/zed/settings.json
{
  "context_servers": {
    "aether": {
      "command": {
        "path": "npx",
        "args": ["-y", "aether-mcp"],
        "env": { "AETHER_API_KEY": "ak_live_a1b2c3d4e5f6g7h8i9j0k1l2m3n4o5p6" }
      }
    }
  }
}

Hard limits

Aether refuses only:

  • CSAM
  • Operational planning of real-world violence against specific identifiable people
  • Synthesis instructions for weapons of mass destruction (CBRN)
  • Active malware deployment against systems you do not own

Everything else — security research, malware analysis, gray-hat web work, controversial research, adult fiction, dark creative writing, blunt opinions, technical questions other AIs hedge on — Aether handles directly.

Pricing

PlanPer-hour rateAPI keys allowedDaily-refresh credits
Free (no card)301150
Starter $15/mo200350
Pro $35/mo1,0005100
Unlimited $99/mounlimited10300

Each call deducts credits based on tokens used — same as web chat. Top up anytime with crypto, no card on file. See full pricing →

Privacy

  • API keys are SHA-256 hashed at rest. We never see your full key after creation.
  • IP and user-agent are logged on every API call for abuse detection — same as the web app.
  • Your prompts and outputs are not sold. Conversations are not used to train third-party models.
  • Crypto-only payment — no name on file, no chargeback risk, anonymous billing.