Atlas / MCP servers / commonhuman-lab / NyxStrike

NyxStrikeBLOCK

Name: NyxStrike
Author: commonhuman-lab

mcp/commonhuman-lab/nyxstrike

MCP server by commonhuman-lab · ai · bug bounty · ctf · Audited 2026-10-07

⚑ Report / re-scan Source ↗

AI Powered penetration testing Platform for offensive security research

Verdict

BLOCK

Grade

Trust score

42 /100

Exposed tools

6 5r · 1w · 0d

Transport

—

License

NOASSERTION

Stars

156

Overview

From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.

AI-powered offensive security orchestration engine

What is NyxStrike?

NyxStrike connects LLM agents to real offensive security tools and executes full attack chains — from recon to exploitation.

200+ tools · 50+ categories · AI decision engine · Tamper-evident evidence chain

🚀 Quick Start (Installation)

Get a full offensive security environment running in minutes.

git clone https://github.com/CommonHuman-Lab/nyxstrike.git
cd nyxstrike

./nyxstrike.sh -a               # Setup + start server
./nyxstrike.sh -a -t            # + install external tools

Full flag reference: Wiki — Installation & Flags

Verify Setup

Open http://localhost:8888 to access the dashboard.

Some tools (e.g. nmap, masscan) require elevated privileges for specific scan modes. Use a dedicated test VM and least-privilege setup where possible.

🔌 AI Agent Integrations (MCP)

Connect NyxStrike to any MCP-compatible AI client — OpenCode, Cursor, Claude Desktop, VS Code Copilot, Roo Code, and more.

Open http://localhost:8888/#/help for help with configurations.

Universal MCP Command

/path/to/nyxstrike/nyxstrike-env/bin/python3 \
/path/to/nyxstrike/nyxstrike_mcp.py \
--server http://127.0.0.1:8888 \
--profile full

OpenCode

{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"nyxstrike": {
"type": "local",
"command": [
"/path/to/nyxstrike/nyxstrike-env/bin/python3",
"/path/to/nyxstrike/nyxstrike_mcp.py",
"--server",
"http://127.0.0.1:8888",
"--profile",
"full"
],
"enabled": true
}
}
}

Config snippets for Claude Desktop, Cursor, VS Code Copilot, and security options: [Wi

Read from source at commit 3f26b9c88a91OBSERVED · 2026-10-07

Connect

Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.

claude-code

claude mcp add nyxstrike --env NYXSTRIKE_API_TOKEN=${NYXSTRIKE_API_TOKEN} --env NYXSTRIKE_CRED_ENC_KEY=${NYXSTRIKE_CRED_ENC_KEY} -- uvx nyxstrike

claude-desktop

{
  "mcpServers": {
    "nyxstrike": {
      "command": "uvx",
      "args": [
        "nyxstrike"
      ],
      "env": {
        "NYXSTRIKE_API_TOKEN": "${NYXSTRIKE_API_TOKEN}",
        "NYXSTRIKE_CRED_ENC_KEY": "${NYXSTRIKE_CRED_ENC_KEY}"
      }
    }
  }
}

Exposed tools (6)

5 read · 1 write · 0 destructive.

Tool	Risk	Description
`classify_task`	read
`nyxstrike_h2csmuggler`	read
`nyxstrike_net_ping`	read
`run_tool`	write
`ssh`	read	try:
`telnet`	read	try:

Trust audit

BLOCKgrade F · trust 42/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.

Layer	What it checks	Result
L0	Provenance & inventory	FAIL
L1	Static analysis of the code	FAIL
L2	Instruction surface (what it tells the agent)	PASS
L3	Class-specific surface	PASS
L4	Behavioural (sandbox)	SKIPPED

What the source does

Filesystem: declared (19 observation(s))
Network: declared (22 observation(s))
Shell: declared (11 observation(s))
Dependencies: not all pinned
Secrets in source: found

Findings (25)

CRITICALObfuscation / stealth · obf.decode_then_exec · CWE-506, CWE-94

backend/server_core/generators/ai_exploit_generator.py:283

b64decode( ... 
exec(

Why it matters. decodes a payload and executes it

CRITICALHard-coded secrets · secret.private_key · CWE-798, CWE-321

frontend/ui/src/app/demo.ts:513

secret: '-----BEGIN OPENSSH PRIVATE KEY-----\nb3BlbnNzaC1rZXktdjEAAAAA...(truncated)\n-----END OPENSSH PRIVATE KEY-----',

HIGHInventory / provenance · inv.suspicious_name · CWE-1104

frontend/ui/src/api/types/payload-workbench.ts

payload-workbench.ts

Why it matters. member named after an attack tool

Fix. remove or justify

HIGHInventory / provenance · inv.suspicious_name · CWE-1104

frontend/ui/src/components/tool-run/payload.ts

payload.ts

Why it matters. member named after an attack tool

Fix. remove or justify

HIGHCode injection · code.eval_exec · CWE-78, CWE-94, CWE-95

mcp_client/registrar.py:76

exec(compile(src, f"<toolspec:{spec.name}>", "exec"), namespace)

Why it matters. evaluates text as code

Fix. remove; use a parser or a dispatch table

HIGHFilesystem / path · fs.credential_store · CWE-22, CWE-59

frontend/dashboard/assets/demo-BVWMV9-U.js:33

-----END OPENSSH PRIVATE KEY-----`,service:`ssh`,host:`10.0.0.55`,port:22,source_tool:`gobuster`,evidence:`Found at https://example.com/backup/.ssh/id_rsa (HTTP 200)`,tags:[`ssh`,`private-key`],verifi

Why it matters. touches a credential store

HIGHFilesystem / path · fs.credential_store · CWE-22, CWE-59

frontend/dashboard/assets/demo-BVWMV9-U.js:42

region = us-east-1`,path:`/home/ubuntu/.aws/credentials`,host:`10.0.0.55`,source_tool:`nuclei`,tags:[`aws`,`cloud`,`credentials`],notes:`Exposed in public S3 bucket listing — rotate immediately`,sessi

Why it matters. touches a credential store

HIGHFilesystem / path · fs.credential_store · CWE-22, CWE-59

frontend/ui/src/app/demo.ts:518

evidence: 'Found at https://example.com/backup/.ssh/id_rsa (HTTP 200)',

Why it matters. touches a credential store

HIGHFilesystem / path · fs.credential_store · CWE-22, CWE-59

frontend/ui/src/app/demo.ts:609

path: '/home/ubuntu/.aws/credentials',

Why it matters. touches a credential store

HIGHNetwork egress · net.metadata_ip · CWE-200, CWE-319

backend/server_core/workflows/bugbounty/workflow.py:155

{"name": "Cloud Metadata", "payloads": ["http://169.254.169.254/latest/meta-data/"]},

Why it matters. cloud metadata endpoint: the classic SSRF credential grab

HIGHNetwork egress · net.tls_off · CWE-200, CWE-319

backend/server_api/web_framework/browser_agent.py:182

resp = requests.get(page_info.get('url', ''), timeout=10, verify=False)

Why it matters. certificate verification is disabled

Fix. leave verification on

HIGHNetwork egress · net.tls_off · CWE-200, CWE-319

backend/server_api/web_framework/browser_agent.py:262

r = requests.get(test_url, timeout=8, verify=False)

Why it matters. certificate verification is disabled

Fix. leave verification on

HIGHNetwork egress · net.tls_off · CWE-200, CWE-319

backend/server_core/tool_specs/web_scan.py:312

argv.append("--insecure")

Why it matters. certificate verification is disabled

Fix. leave verification on

HIGHNetwork egress · net.tls_off · CWE-200, CWE-319

mcp_client/api_client.py:24

self.session.verify = False  # Disable SSL verification for self-signed certs

Why it matters. certificate verification is disabled

Fix. leave verification on

MEDIUMCode injection · code.dynamic_import · CWE-78, CWE-94, CWE-95

backend/server_core/payload_workbench/registry.py:73

category_pkg = importlib.import_module(f"backend.server_core.payload_workbench.operations.{category_name}")

MEDIUMCode injection · code.dynamic_import · CWE-78, CWE-94, CWE-95

backend/server_core/workbench/registry.py:76

category_pkg = importlib.import_module(f"backend.server_core.workbench.operations.{category_name}")

MEDIUMCode injection · code.dynamic_import · CWE-78, CWE-94, CWE-95

mcp_client/registrar.py:84

module = importlib.import_module(f"backend.server_core.tool_specs.{category}")