Compuute MCP Security Scanner is an MCP server published by io.github.Compuute, hosted at scan.compuute.se. When last checked on 2026-11-21 it was working: it answers and lists its tools. It declares 1 tool, including scan_mcp_server.
Scan any public GitHub MCP-server repo for security issues. 37 MCP-specific L1 rules, 8 languages.
10 connections through 5 shared values. Tap any node.
Inferred from the host name and headers: what a server runs on, not who runs it.
Scan a public GitHub MCP-server repository for security issues. Clones the repo (shallow, <60s, <200 MB), runs compuute-scan v0.6.2 in static analysis mode (no code execution from the target), and returns a structured report with severity counts, a 0-100 score, and the 10 most severe findings. WHEN TO USE: - Before connecting to an unknown MCP server discovered via Anthropic Registry, Smithery, mcp.so, or a Discord recommendation. - Before installing a third-party MCP-server package into a production pipeline. - As part of an agent's pre-commit / pre-deploy due-diligence step when adding new dependencies. - As one input to a multi-source trust evaluation (combine with publisher reputation, package install count, last-update recency). WHEN NOT TO USE: - For private repos. Use the on-prem CLI instead: `npx compuute-scan ./path-to-private-repo` - For deep exploitability assessment of a specific code path. This is pattern matching, not dataflow analysis. Book a manual L2-L4 audit at https://compuute.se/audit for that depth. - For non-GitHub hosts (GitLab, Bitbucket, self-hosted). v1 supports github.com only. - For repos > 200 MB or clone time > 60s. The endpoint returns a 413 or 504 in those cases — fall back to local CLI. EXPECTED RESPONSE TIME: - Median: ~1-2 seconds for small repos (<100 files). - p99: ~10 seconds for medium repos. - Hard timeout at clone=60s, scan=120s combined. EXPECTED COST: - Free tier in MVP. Future Pro tier may charge per-scan or per-month. DATA FRESHNESS: - Scanner version is reported in response.scanner.version. - L1 rule set freshness reflects compuute-scan releases — see github.com/Compuute/compuute-scan/CHANGELOG.md for the latest CVE and threat-intel response timeline. EXAMPLES: Example 1 — scan an MCP server you're evaluating: github_url = "https://github.com/modelcontextprotocol/servers" → score: 0, summary: {critical: 1, high: 94, medium: 22} → top_findings include SSRF, eval, etc. → recommendation: "AVOID — 1 critical and 94 high finding(s)..." Example 2 — scan a clean reference implementation: github_url = "https://github.com/microsoft/azure-devops-mcp" → score: 90+, summary: {critical: 0, high: 1} → recommendation: "REVIEW — 1 high finding(s)..." Example 3 — scan your own dev MCP-server before publishing: github_url = "https://github.com/yourorg/your-mcp" → audit your own surface before others install it OUTPUT FIELDS (stable schema): - repo_url (str): canonical URL of the scanned repo. - score (int): 0-100, higher safer. Coarse summary, not a precision claim. - summary (object): {critical, high, medium, low, info, files_scanned}. - recommendation (str): action guidance derived from severity counts. - findings_count (int): total raw findings (may include false positives). - top_findings (list): up to 10 most severe, each with {id, title, severity, file, line, owasp, cwe}. - l0_discovery (object): MCP transport, tool count, dependency pinning. - performance (object): clone_seconds, scan_seconds, repo_size_bytes. - scanner (object): {name, version, layers_covered}. - _disclaimer (str): MANDATORY triage disclaimer. Read it. Args: github_url: Public GitHub HTTPS URL (e.g. https://github.com/org/repo). Must be public and < 200 MB. v1 is github.com only. Returns: Structured scan result. On error, returns {"error": code, "message": ...} with HTTP-style code (invalid_url, clone_failed, scan_timeout, etc.).
None yet. Changes appear after a server's second daily check.
Squares are this server's values; dots are other servers sharing them. A shared address or publisher is a lead, not proof of a common operator.
| Parameter | Type | Description |
|---|---|---|
| github_url * | string |
{
"name": "scan_mcp_server",
"description": "Scan a public GitHub MCP-server repository for security issues.\n\n Clones the repo (shallow, <60s, <200 MB), runs compuute-scan v0.6.2 in\n static analysis mode (no code execution from the target), and returns a\n structured report with severity counts, a 0-100 score, and the 10 most\n severe findings.\n\n WHEN TO USE:\n - Before connecting to an unknown MCP server discovered via Anthropic\n Registry, Smithery, mcp.so, or a Discord recommendation.\n - Before installing a third-party MCP-server package into a production\n pipeline.\n - As part of an agent's pre-commit / pre-deploy due-diligence step\n when adding new dependencies.\n - As one input to a multi-source trust evaluation (combine with\n publisher reputation, package install count, last-update recency).\n\n WHEN NOT TO USE:\n - For private repos. Use the on-prem CLI instead:\n `npx compuute-scan ./path-to-private-repo`\n - For deep exploitability assessment of a specific code path. This is\n pattern matching, not dataflow analysis. Book a manual L2-L4 audit\n at https://compuute.se/audit for that depth.\n - For non-GitHub hosts (GitLab, Bitbucket, self-hosted). v1 supports\n github.com only.\n - For repos > 200 MB or clone time > 60s. The endpoint returns a 413\n or 504 in those cases — fall back to local CLI.\n\n EXPECTED RESPONSE TIME:\n - Median: ~1-2 seconds for small repos (<100 files).\n - p99: ~10 seconds for medium repos.\n - Hard timeout at clone=60s, scan=120s combined.\n\n EXPECTED COST:\n - Free tier in MVP. Future Pro tier may charge per-scan or per-month.\n\n DATA FRESHNESS:\n - Scanner version is reported in response.scanner.version.\n - L1 rule set freshness reflects compuute-scan releases — see\n github.com/Compuute/compuute-scan/CHANGELOG.md for the latest CVE\n and threat-intel response timeline.\n\n EXAMPLES:\n\n Example 1 — scan an MCP server you're evaluating:\n github_url = \"https://github.com/modelcontextprotocol/servers\"\n → score: 0, summary: {critical: 1, high: 94, medium: 22}\n → top_findings include SSRF, eval, etc.\n → recommendation: \"AVOID — 1 critical and 94 high finding(s)...\"\n\n Example 2 — scan a clean reference implementation:\n github_url = \"https://github.com/microsoft/azure-devops-mcp\"\n → score: 90+, summary: {critical: 0, high: 1}\n → recommendation: \"REVIEW — 1 high finding(s)...\"\n\n Example 3 — scan your own dev MCP-server before publishing:\n github_url = \"https://github.com/yourorg/your-mcp\"\n → audit your own surface before others install it\n\n OUTPUT FIELDS (stable schema):\n - repo_url (str): canonical URL of the scanned repo.\n - score (int): 0-100, higher safer. Coarse summary, not a precision claim.\n - summary (object): {critical, high, medium, low, info, files_scanned}.\n - recommendation (str): action guidance derived from severity counts.\n - findings_count (int): total raw findings (may include false positives).\n - top_findings (list): up to 10 most severe, each with {id, title,\n severity, file, line, owasp, cwe}.\n - l0_discovery (object): MCP transport, tool count, dependency pinning.\n - performance (object): clone_seconds, scan_seconds, repo_size_bytes.\n - scanner (object): {name, version, layers_covered}.\n - _disclaimer (str): MANDATORY triage disclaimer. Read it.\n\n Args:\n github_url: Public GitHub HTTPS URL (e.g. https://github.com/org/repo).\n Must be public and < 200 MB. v1 is github.com only.\n\n Returns:\n Structured scan result. On error, returns {\"error\": code, \"message\": ...}\n with HTTP-style code (invalid_url, clone_failed, scan_timeout, etc.).\n ",
"inputSchema": {
"type": "object",
"title": "scan_mcp_serverArguments",
"required": [
"github_url"
],
"properties": {
"github_url": {
"type": "string",
"title": "Github Url"
}
}
}
}Scan a public GitHub MCP-server repo with compuute-scan, the MCP-specific static security scanner (38 L1 rules across TS/JS, Python, Go, Rust, C#, Java, Kotlin). Use this before an agent connects to an unknown MCP server or before installing a third-party MCP-server package. Every response carries a triage disclaimer: findings are pattern matches, not exploitability claims — manual dataflow review required for production decisions.
Names, descriptions and schemas were written by the server and are shown as data.
Shared addresses on big platforms (Cloudflare, Vercel and the like) put unrelated servers together; those are marked shared_infrastructure. A match is a pivot, not a link between operators.