Sn1per MCP Server: pentest, recon and attack surface scanning runs on your own machine — the client starts it, so there is no endpoint to ping. 111 installs a week from pypi. Last commit 17 Oct 2026.
Run scoped, authorized Sn1per pentest, recon and attack surface scans from AI agents; read findings.
We read the source, 2 d ago · rules 995398eba69f
What this server is able to do. For an MCP server this is often the job itself — a terminal server runs commands because that is what it is for. Listed so you know what you are plugging in, not as an accusation.
return os.path.join(self.state_dir, state.JOBS_SUBDIR)
return subprocess.Popen(
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
We found places where it runs commands, builds paths or queries from values it is given. None of that is a flaw by itself — it becomes one when the code changes, and code changes quietly between releases. We re-read it on every one.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add sn1per-mcp -- uvx sn1per-mcp
{
"mcpServers": {
"sn1per-mcp": {
"command": "uvx",
"args": [
"sn1per-mcp"
]
}
}
}
[mcp_servers.sn1per-mcp]
command = "uvx"
args = ["sn1per-mcp"]
{
"mcpServers": {
"sn1per-mcp": {
"command": "uvx",
"args": [
"sn1per-mcp"
]
}
}
}
{
"servers": {
"sn1per-mcp": {
"type": "stdio",
"command": "uvx",
"args": [
"sn1per-mcp"
]
}
}
}
Optional settings from the registry entry, left out of the block above:
SNIPER_MCP_SCOPE (Comma-separated domains, IPs and CIDR ranges you are authorized to test (prefix ! to exclude). Scans outside it are refused), SNIPER_MCP_SCOPE_FILE (Optional absolute path to a scope file with one or more entries per line and # comments), SNIPER_MCP_AUTHORIZATION_REF (Optional authorization reference, such as a statement-of-work or ticket id, recorded in the audit log), SNIPER_MCP_ALLOW_INTRUSIVE (Set to true to offer scan modes that run brute force, exploitation or injection testing. Off by default), SNIPER_MCP_ENABLE_DESTRUCTIVE (Set to true to add the workspace delete and export tools; they ask the human to confirm. Off by default), SNIPER_MCP_REDACT (Redaction of secrets in results. On by default; turn off only with a local model when raw values are required), SNIPER_EDITION (Optional override of the auto-detected Sn1per edition), SNIPER_BIN (Optional absolute path to the sniper CLI. Detected automatically on standard installs), SNIPER_INSTALL_DIR (Optional Sn1per install directory. Read from the CLI or /usr/share/sniper by default), SNIPER_LOOT_BASE (Optional directory that contains the loot workspace tree. Defaults to the install directory's loot folder), SNIPER_SCAN_ENABLED (Set to false for a read-only deployment that never launches scans).
Drive ASMHunter attack-surface monitoring from an AI agent: recon, scans, findings, reports.
MCP server for Pentest-Tools.com: run scans, manage findings and reports via your preffered LLM.
MCP server for Sublime Text 4. Lets AI coding agents read and control a running ST instance.
Start index365 scans and read prioritized findings your coding agent can use.
MCP server for t-rush. Exposes TODO/FIXME scanning and streak tracking to AI agents.
Attack paths, scan findings and shadow AI for AI-native apps and their clouds; vet MCP servers.
Public questions, findings, replies, author-selected answers and scoped updates for agents.
Authorization-gated MCP server to discover and run 670+ security tools for CTF, pentest, and DFIR.
Answers built from our own checks of this server.