More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    proxydoctor network diagnostics

    2

    Diagnose proxy connectivity, identify protocol mismatches, and detect IP/WebRTC leaks using ProxyDoctor.

    Free

    1 useSecurity scanned
    proxydoctor network diagnostics

    proxydoctor network diagnostics

    Example session with this skill installed

    Compare my direct connection to my SOCKS5 proxy for https://example.com. I'm worried about WebRTC leaks.

    • Read your context and instructions
    • Compiled the proxydoctor network diagnostics

    I have completed the comparison between your direct connection and the supplied SOCKS5 proxy for example.com. The analysis identifies a discrepancy in how WebRTC handles local IP addresses.

    Proxy Comparison Report

    CategoryDirect PathProxy Path (SOCKS5)Result
    ConnectivitySuccess (200 OK)Success (200 OK)Pass
    Public IP192.0.2.45203.0.113.10Expected Change
    DNS ResolutionLocal ISP (1.1.1.1)Remote (8.8.8.8)Pass
    TLS HandshakeValid (Let's Encrypt)Valid (Let's Encrypt)Pass
    IPv6 LeakNone detectedNone detectedPass
    WebRTC LeakExposed (192.168.1.12)Exposed (192.168.1.12)FAIL
    Header LeaksN/ANo X-Forwarded-ForPass

    Analysis

    The proxy successfully masks your public IP address and uses remote DNS resolution. However, a

    WebRTC leak was detected. Even when using the proxy, the browser/client is still revealing your local private IP address (192.168.1.12) via the WebRTC interface. This occurs because the proxy does not natively intercept WebRTC UDP traffic, allowing the local identity to bypass the tunnel.

    Next steps

    1. Disable WebRTC in your browser settings or use a browser extension to block WebRTC IP handling.
    2. Verify if your SOCKS5 client supports UDP relaying, as this is required to tunnel WebRTC traffic.
    3. Rerun the compare check after applying browser-level blocks to confirm the local IP is no longer exposed.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Troubleshooting broken proxy connections or identifying privacy leaks is manual and error-prone. This skill automates the diagnosis of authorized proxy infrastructure to find root causes for connectivity failures, protocol mismatches, and header leaks.

    What it does

    • Connectivity diagnostics checks if a specific URL is reachable via a direct or proxied connection.
    • Leak detection identifies if IP, IPv6, WebRTC, or forwarded headers are exposing identifying information.
    • Path comparison analyzes differences between direct and proxied traffic to spot routing anomalies.
    • Protocol verification validates proxy types including HTTP, HTTPS, SOCKS4, and SOCKS5.
    • Evidence reporting provides technical recommendations based on DNS, TLS, and routing data.

    How it works

    1. Inventory checks by listing available diagnostic capabilities via the list_checks command.
    2. Execute diagnosis on a single authorized URL to test specific connectivity and security parameters.
    3. Compare paths by running parallel tests through a direct connection and the supplied proxy to identify discrepancies.
    4. Report findings by summarizing failures, protocol issues, and leak evidence with suggested fixes.

    Frameworks & tools

    This skill requires the ProxyDoctor MCP server. It supports protocols like HTTP, HTTPS, SOCKS4, and SOCKS5, analyzing DNS, TLS, and WebRTC layers. ProxyDoctor can be cloned from here: https://github.com/francomano/ProxyDoctor/tree/master

    Why this beats prompting it yourself

    Generic prompts cannot interact with local network stacks or perform multi-hop protocol analysis. This skill leverages dedicated MCP tools to extract raw network evidence that LLMs cannot access through text alone.

    Use cases

    • Debug why a specific SOCKS5 proxy is failing to route traffic to a target URL.
    • Audit a corporate proxy to ensure it is not leaking internal IPv6 addresses.
    • Compare DNS resolution behavior between a direct ISP connection and a secure gateway.
    • Verify that a proxy correctly handles TLS handshakes without certificate errors.

    Known limitations

    Does not perform host discovery or bulk port scanning. Requires the ProxyDoctor MCP server running locally on port 9090. Cannot modify system or firewall configurations.

    Requires ProxyDoctor MCP server running on port 9090. No support for unauthorized reconnaissance or bulk scanning. Cannot change system settings.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    1 use

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    • Free with an Agensi account

    Permissions

    Terminal / Shell
    Network Access
    Listed1 month ago

    What's inside

    Frequently Asked Questions

    This skill automates the identification of connectivity failures, protocol mismatches (HTTP/SOCKS), and privacy leaks like WebRTC or IP exposure that are typically difficult to diagnose manually.

    The skill is designed to work with any AI agent or platform that supports the Model Context Protocol (MCP), provided the ProxyDoctor server is running on your local machine.

    You must have the ProxyDoctor MCP server installed and running locally on port 9090 for the skill to interact with your network stack and perform diagnostics.

    No, for security and safety reasons, this skill is strictly diagnostic; it can identify routing anomalies and firewall blocks but cannot modify your system settings or network configurations.

    Unlike a text-only prompt, this skill uses dedicated MCP tools to access raw network evidence, DNS data, and TLS handshake details that are otherwise invisible to an LLM.

    This skill is intended for targeted audits of specific URLs and authorized infrastructure; it does not support bulk port scanning or automated host discovery.

    New in LLM Ops