MCP/Risk

Directory / server

ai.flowcastle/flowcastle

registry.modelcontextprotocol.io, scanned 2026-09-11, tool surface via remote-tools-list

This directory entry reports an automated security scan of ai.flowcastle/flowcastle, scanned 2026-09-11. Trust grade C rates provenance, transport, known vulnerabilities and instruction integrity; capability risk is reported separately, not folded into the grade.

Tool surface changed. This server's tool schema differs from its previous scan (2026-07-28). Re-review §2 and §5 before continuing to trust it.

Automated checks

StatusCheckPolicyResult
PASS Listed on the official MCP registry §1.1
Listed on registry.modelcontextprotocol.io as ai.flowcastle/flowcastle. 1 finding
  • Registry entry: ai.flowcastle/flowcastle ↗
PASS Source repository is public and maintained §1.2
Active repository with a license. 5 findings
  • Repository ↗
  • Last push: 47 days ago
  • Stars: 2
  • License: MIT
  • Security policy (SECURITY.md): absent
INFO Published package is consistent and attested §1.3

No published package associated with this server (remote-only or non-npm/PyPI source).
PASS Tool surface is inspectable and proportionate §2.1–§2.4
50 tool(s) enumerated. Capabilities detected: network-egress, filesystem. Reported as capability risk (§2.4); whether each is essential to the server's stated purpose is §2.3 and remains a manual judgement. 4 findings
  • Tools exposed: 50
  • get_block_details: network-egress
  • apply_actions: filesystem
  • Detected capabilities: network-egress, filesystem
PASS HTTPS-only endpoint §3.1
Endpoint uses HTTPS. 1 finding
  • Endpoint: https://api.flowcastle.ai/api/mcp
FAIL Authentication required §3.2

Server answers anonymous requests (HTTP 200) while exposing non-trivial or unverifiable capabilities - fail per §3.2.
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://api.flowcastle.ai/.well-known/oauth-protected-resource
INFO No known vulnerabilities (OSV.dev) §4.1

No package to query (remote-only server).
WARN Server-supplied instructions free of poisoning indicators §5.1–§5.4
2 suspicious indicator(s) found in server-supplied instructions, tool descriptions, or schemas. 2 findings
  • apply_actions description: oversized: 2194 chars (limit 2048)
  • send_message description: oversized: 2192 chars (limit 2048)

Score history

DateGradeScore
2026-09-11C65/100
2026-07-28B88/100
2026-07-27B88/100
2026-07-27B88/100
2026-07-26B88/100
2026-07-26B88/100
2026-07-26B88/100
2026-07-25D61/100
2026-07-25D61/100
2026-07-25D57/100
2026-07-25B84/100
2026-07-24A100/100
2026-07-24B88/100
2026-07-24B89/100
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.