MCP Risk

Directory / server

ai.borealhost/mcp

This directory entry reports an automated security scan of ai.borealhost/mcp, scanned 2026-07-29. Trust grade C rates provenance, transport, known vulnerabilities and instruction integrity; capability risk is reported separately, not folded into the grade.

registry.modelcontextprotocol.io, scanned 2026-07-29, tool surface via remote-tools-list

C 65/100 permalink

Automated checks

StatusCheckPolicyResult
PASS Listed on the official MCP registry §1.1
Listed on registry.modelcontextprotocol.io as ai.borealhost/mcp. 1 finding
  • Registry entry: ai.borealhost/mcp ↗
WARN Source repository is public and maintained §1.2
Repository health issues: no license. 5 findings
  • Repository ↗
  • Last push: 105 days ago
  • Stars: 0
  • License: none detected
  • Security policy (SECURITY.md): absent
FAIL Published package is consistent and attested §1.3
Integrity concerns: package repository field does not match the claimed source repo; no PyPI trusted-publishing provenance. 5 findings
  • Package: borealhost-mcp@0.2.0 ↗
  • Ecosystem: PyPI
  • Age: 107 days
  • Repo field matches source repo: false
  • Trusted-publishing provenance: absent
PASS Tool surface is inspectable and proportionate §2.1–§2.4
104 tool(s) enumerated. Capabilities detected: filesystem, network-egress. Reported as capability risk (§2.4); whether each is essential to the server's stated purpose is §2.3 and remains a manual judgement. 8 findings
  • Tools exposed: 104
  • list_files: filesystem
  • write_file: filesystem
  • upload_file: filesystem
  • delete_file: filesystem
  • create_alert_rule: network-egress
  • get_logs: network-egress
  • Detected capabilities: filesystem, network-egress
PASS HTTPS-only endpoint §3.1
Endpoint uses HTTPS. 1 finding
  • Endpoint: https://borealhost.ai/mcp/
UNVERIFIABLE Authentication required §3.2

Endpoint rejected the probe (HTTP 400); whether it requires authentication could not be determined.
WARN OAuth protected-resource metadata published §3.3
No protected-resource metadata found - clients cannot discover the authorization server per spec. 1 finding
  • Checked: https://borealhost.ai/.well-known/oauth-protected-resource
PASS No known vulnerabilities (OSV.dev) §4.1

No advisories on record for borealhost-mcp @0.2.0 (PyPI).
WARN Server-supplied instructions free of poisoning indicators §5.1–§5.4
3 suspicious indicator(s) found in server-supplied instructions, tool descriptions, or schemas. 3 findings
  • server instructions: oversized: 6377 chars (limit 4096)
  • get_billing_portal: embedded URL: https://billing.stripe.com/p/session/...
  • database_search_replace: embedded URL: http://old-domain.com

Score history

DateGradeScore
2026-07-29C65/100
2026-07-27D58/100
2026-07-27D58/100
2026-07-27D58/100
2026-07-27D58/100
2026-07-26D58/100
2026-07-26D58/100
2026-07-26D58/100
2026-07-26F23/100
2026-07-25F25/100
2026-07-25F25/100
2026-07-25F25/100
2026-07-25F25/100
2026-07-24F18/100
2026-07-24F18/100
2026-07-24F16/100
2026-07-12F22/100
What this grade covers
  • Provenance & supply-chain integrity (§1)
  • Tool surface & capability risk (§2)
  • Transport & authentication (§3)
  • Known CVEs via OSV.dev (§4)
  • Instruction/tool poisoning patterns (§5)

Limitations: pattern-based only (no semantic injection detection), static npm analysis (live server instructions not retrieved for package-only scans), single-server scope (no cross-server toxic flow analysis). See the full policy.

Methodology: MCP Server Security Policy. Unfamiliar term? See the glossary.