correctover-scan MCP server
Commit Nov 7, 2026Audits MCP configurations and AI agent bundles for security risks like credential leaks, SSRF, and command injection.
Also known as DSHCorrectover/correctover-scan
- Stars
- 1
- Tools
- —
- Downloads, 30 days
- —
- Latest version
- —
- Last commit
- Nov 7, 2026
- License
- MIT
What is correctover-scan MCP?
correctover-scan audits MCP configurations and AI agent bundles for security risks like credential leaks, SSRF, and command injection.
Summary generated from the project's README
About correctover-scan MCP server
Auditing MCP configurations and AI agent bundles for security vulnerabilities is the primary function of correctover-scan. Developers using Claude Desktop, Claude Code, Cursor, or VS Code can use this tool to identify risks such as credential leaks, SSRF exposure, command injection, and missing authentication. It operates locally to ensure configurations are never uploaded, making it a private way to check for potential remote code execution or data exfiltration paths.
The scanner follows the CCS-aligned methodology and includes 14 checks mapped to OWASP AISVS 1.0. For those working with published npm packages or minified JavaScript, the bundle mode provides 17 specific checks that return findings with file and line numbers. Users can run the tool immediately without a complex installation process by using the npx correctover-scan command. It also supports CI environments and generates SARIF files for integration with GitHub code scanning.
What you can do with it
- 1Audit local agent configurationsCheck your Claude Desktop or Cursor MCP settings for security flaws like exposed credentials.
- 2Scan npm package bundlesPerform a security scan on published npm packages or minified JavaScript files using bundle mode.
- 3Integrate security into CI/CDAutomate configuration audits within your development pipeline using SARIF output.
Written from the project's README. Every use case is checked against a quote from it.
README
View on GitHubcorrectover-scan
CCS-aligned security scanner for MCP servers and AI agents (IETF draft-correctover-ccs). Credential exposure, SSRF, command-injection risk, missing auth. 14 checks → OWASP AISVS 1.0. Zero install:
npx correctover-scan.
🔒 We Practice What We Preach
We scan our own code with correctover-scan every week. Latest self-audit found real vulnerabilities (SSRF, missing input validation) — and we fixed them publicly. Read our self-audit report →
Current Security Score: 76/100 (improving weekly, target: 95+/100)
Scan finds risk in config. Runtime blocks the call.
After scanning, put enforcement on the path:
npx -y agent-runtime-guard # fail-closed tool-call intercept + Ed25519 receipts
→ agent-runtime-guard · COMPARE.md
Verifiable artifacts behind this tool
- Third-party interoperability — joint assessment merged into the EMILIA protocol
- 66 signed conformance test vectors — ccs-conformance-vectors
- Published methodology — Zenodo DOI 10.5281/zenodo.21783723
- Human audit — 116-check methodology: Agent Output Audit
correctover-scan audits MCP configuration used by Claude Code, Claude Desktop, Cursor, VS Code and other agents. It flags credential leaks, SSRF exposure, missing TLS, over-broad tools and related issues — the config problems that become RCE, exfil and injection paths. Runs locally (no upload), CI-friendly, SARIF for GitHub code scanning.
v1.4.0 adds --bundle mode: signal scanning over published npm packages and minified JS (17 checks, file+line findings).
Where it sits (vs other tools)
| Layer | Tool | Job |
|---|---|---|
| Config scan (this repo) | correctover-scan | Find risky MCP / agent config before runtime |
| Runtime intercept | agent-runtime-guard | Block this tool call; emit receipt |
| Egress firewall | Pipelock-class | Mediate outbound network |
| Discovery / posture | Reco / Pluto-class | Inventory agents & permissions |
| Ecosystem data | mcp-atlas-census | 229,827 tools with schema (CC0) |
Scanners do not stop a live curl|sh. Runtime guards do not replace a full config audit. Use both.
⚡ Quick Start — 30 seconds
npx correctover-scan@latest
Auto-detects .cursor/mcp.json, claude_desktop_config.json, .claude/mcp.json, mcp.json, .vscode/mcp.json, and more. Local only — configs are not uploaded.
correctover-scan MCP questions
What security risks does the scanner detect?
The tool identifies credential exposure, SSRF, command-injection risk, and missing authentication.
Does the scanner upload my configuration files?
No, the scanner runs locally and does not upload your configurations.
Which AI agent configurations can be audited?
It audits configurations used by Claude Code, Claude Desktop, Cursor, VS Code, and other agents.
Can I scan published npm packages?
Yes, the --bundle mode allows for scanning over published npm packages and minified JS.