AI agents invoke start_php_server to trigger actions in Pentester-MCP. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
This tool starts a PHP web server, which executes arbitrary PHP code at runtime. While not inherently destructive, it enables execution of code and external operations—core characteristics of the Execute category. The high severity reflects that a misconfigured or malicious PHP server started by an AI agent could be exploited to compromise systems or host malicious payloads.
From the tool's definition Tool name 'start_php_server' combined with server context describing 'penetration testing tools' and 'web exploitation'; starting a PHP server is a code execution operation that launches an external service whose behavior depends on the runtime environment…
Documented attack patterns abuse exactly the kind of access start_php_server gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Pentester-MCP, and nothing reaches the server without passing your rules. This is the rule we recommend for start_php_server:
{
"version": "1",
"default": "deny",
"tools": {
"start_php_server": {
"limits": [
{
"counter": "start_php_server_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}start_php_server stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
start_php_server. It is categorised as a Execute tool in the Pentester-MCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Pentester- MCP server in PolicyLayer and add a rule for start_php_server: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Pentester-MCP. Nothing to install.
start_php_server is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the start_php_server rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for start_php_server. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
start_php_server is provided by the Pentester- MCP server (halilkirazkaya/pentester-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Pentester-MCP, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
337 Pentester-MCP tools catalogued and risk-classified — across an index of 43,000+ MCP servers.