Security audit CLI for MCP (Model Context Protocol) configurations. Find leaked API keys, excessive permissions, and vulnerable servers before attackers do.
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"mcp-vigil","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# 🛡️ mcp-guard **Security audit CLI for MCP (Model Context Protocol) configurations.** Find leaked API keys, excessive permissions, and vulnerable servers before attackers do. ```bash pip install mcp-vigil mcp-vigil scan ``` --- ## Demo ``` $ mcp-vigil scan ╔══════════════════════════════════════╗ ║ MCP GUARD — Security Audit ║ ╚══════════════════════════════════════╝ 📄 ~/.cursor/mcp.json Servers: 5 | Findings: 0 critical, 3 high, 1 other | Score: 67/100 (C) 🟠 [HIGH] Excessive filesystem access: / Server 'filesystem' has --directory set to / — entire disk exposed. Fix: Restrict to a specific subdirectory. 🟠 [HIGH] Secret in env block of MCP config Server 'github', env: GITHUB_PERSONAL_ACCESS_TOKEN Contains a hardcoded secret value in the config file. Fix: Move to .env file. 🟠 [HIGH] Secret in env block of MCP config Server 'slack', env: SLACK_BOT_TOKEN Contains a hardcoded secret value in the config file. Fix: Move to .env file. 🟡 [MEDIUM] MCP server using HTTP without TLS Server 'insecure-remote' connects via http:// — data not encrypted. Fix: Use https:// instead. ──────────────────────────────────────── Files scanned: 1 Total findings: 4 (3 HIGH, 1 MEDIUM) ⚠️ High-severity issues found — fix soon ```  --- ## What it does Scans your MCP configuration files and finds: - 🔴 **Hardcoded API keys** — GitHub tokens, OpenAI keys, AWS credentials in plain text - 🟠 **Excessive filesystem access** — Servers with access to `/` or `/home` - 🟠 **Untrusted server sources** — MCP servers from pastebin, raw GitHub gists, etc - 🟠 **Known CVEs** — Checks against CVE-2025-49596, CVE-2025-6514, and more - 🟡 **Unpinned versions** — npx packages without version pins (supply chain risk) - 🟡 **Insecure transport** — HTTP without TLS for remote servers - 🔵 **Missing environment variables** — Referenced bu…
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.