Raw scan report

Reflect — the complete, unedited output of the deterministic mcptrustchecker engine v1.13.0, scanned . Every finding, capability tag and score component below is exactly what the engine produced — no AI, no post-processing.

← Back to the scan page
{
  "tool": {
    "name": "mcptrustchecker",
    "version": "1.13.0",
    "methodologyVersion": "mcptrustchecker-1.13"
  },
  "target": {
    "id": "reflect-mcp",
    "source": {
      "kind": "package",
      "origin": "reflect-mcp"
    },
    "server": {
      "name": "reflect-mcp"
    }
  },
  "grade": "A",
  "score": {
    "score": 92,
    "threatScore": 100,
    "grade": "A",
    "band": "A",
    "categorySubtotals": {
      "injection": 0,
      "exfiltration": 0,
      "permissions": 0,
      "supply-chain": 0,
      "network": 0,
      "hygiene": 0
    },
    "vector": [
      {
        "kind": "client",
        "term": "capability-exposure",
        "level": "high",
        "label": "capability blast radius (high) — client exposure if the model is manipulated",
        "appliedPenalty": 6
      },
      {
        "kind": "client",
        "term": "verification-discount",
        "level": "none",
        "label": "publisher verification (unlinked) — no provenance/repo link, but the shipped source was fully read",
        "appliedPenalty": 2
      },
      {
        "kind": "client",
        "term": "coverage-honesty",
        "level": "source",
        "label": "inspection depth (source) — how much of the target the scan could see",
        "appliedPenalty": 0
      }
    ],
    "gatesFired": [],
    "methodologyVersion": "mcptrustchecker-1.13"
  },
  "capability": {
    "level": "high",
    "reasons": [
      "can execute shell commands or code"
    ],
    "tags": [
      "code-exec"
    ]
  },
  "coverage": {
    "level": "source",
    "inputs": {
      "toolSurface": true,
      "implementationSource": true,
      "packageMetadata": true,
      "liveTransport": false
    },
    "caveats": [
      "Tools were statically extracted from the published source (12 recovered), not enumerated from a running server. Tool-poisoning, Unicode-smuggling, capability and toxic-flow analysis ran on this inferred surface, but a mis-parsed registration could be missed or mis-attributed, so tool-derived findings are capped below “confirmed”. To grade the real runtime surface, scan the running server: --command \"npx -y <package>\"."
    ]
  },
  "findings": [
    {
      "ruleId": "MTC-SRC-001",
      "title": "Dynamic code execution in server code (reflect_mcp/lib/sanitize.py)",
      "category": "permissions",
      "severity": "high",
      "confidence": "strong",
      "description": "In the server's implementation (`reflect_mcp/lib/sanitize.py:554`): Evaluating strings as code is the most direct RCE primitive; if any tool input reaches it, the server executes attacker-chosen code. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
      "remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
      "location": {
        "kind": "server",
        "name": "reflect_mcp/lib/sanitize.py"
      },
      "evidence": "\"\"\" Run Layer 2 eval (with cache). Returns (result, from_cache: bool). On API failure returns (result with a",
      "owasp": "LLM05:2025 Improper Output Handling",
      "data": {
        "rule": "MTC-SRC-001",
        "file": "reflect_mcp/lib/sanitize.py",
        "line": 554,
        "nonRuntime": false
      }
    },
    {
      "ruleId": "MTC-SRC-002",
      "title": "Shell/command execution in server code (reflect_mcp/cli/docs.py)",
      "category": "permissions",
      "severity": "high",
      "confidence": "strong",
      "description": "In the server's implementation (`reflect_mcp/cli/docs.py:45`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
      "remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
      "location": {
        "kind": "server",
        "name": "reflect_mcp/cli/docs.py"
      },
      "evidence": "less\") try: subprocess.run([pager, str(doc_path)]) except FileNotFoundError: # Fallback: just pr",
      "owasp": "LLM05:2025 Improper Output Handling",
      "data": {
        "rule": "MTC-SRC-002",
        "file": "reflect_mcp/cli/docs.py",
        "line": 45,
        "nonRuntime": false
      }
    },
    {
      "ruleId": "MTC-SRC-002",
      "title": "Shell/command execution in server code (reflect_mcp/cli/doctor.py)",
      "category": "permissions",
      "severity": "high",
      "confidence": "strong",
      "description": "In the server's implementation (`reflect_mcp/cli/doctor.py:65`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
      "remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
      "location": {
        "kind": "server",
        "name": "reflect_mcp/cli/doctor.py"
      },
      "evidence": "try: out = subprocess.run( [\"crontab\", \"-l\"], capture_output=True, text",
      "owasp": "LLM05:2025 Improper Output Handling",
      "data": {
        "rule": "MTC-SRC-002",
        "file": "reflect_mcp/cli/doctor.py",
        "line": 65,
        "nonRuntime": false
      }
    },
    {
      "ruleId": "MTC-SRC-002",
      "title": "Shell/command execution in server code (reflect_mcp/cli/main.py)",
      "category": "permissions",
      "severity": "high",
      "confidence": "strong",
      "description": "In the server's implementation (`reflect_mcp/cli/main.py:1307`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
      "remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
      "location": {
        "kind": "server",
        "name": "reflect_mcp/cli/main.py"
      },
      "evidence": "continue r = subprocess.run( [\"sudo\", \"chattr\", \"+i\", str(path)], capture_output=True",
      "owasp": "LLM05:2025 Improper Output Handling",
      "data": {
        "rule": "MTC-SRC-002",
        "file": "reflect_mcp/cli/main.py",
        "line": 1307,
        "nonRuntime": false
      }
    },
    {
      "ruleId": "MTC-SRC-002",
      "title": "Shell/command execution in server code (reflect_mcp/cli/ops/init.py)",
      "category": "permissions",
      "severity": "high",
      "confidence": "strong",
      "description": "In the server's implementation (`reflect_mcp/cli/ops/init.py:447`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
      "remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
      "location": {
        "kind": "server",
        "name": "reflect_mcp/cli/ops/init.py"
      },
      "evidence": "None: try: subprocess.run( [\"git\", \"init\", str(project_dir)], check=True,",
      "owasp": "LLM05:2025 Improper Output Handling",
      "data": {
        "rule": "MTC-SRC-002",
        "file": "reflect_mcp/cli/ops/init.py",
        "line": 447,
        "nonRuntime": false
      }
    },
    {
      "ruleId": "MTC-SRC-002",
      "title": "Shell/command execution in server code (reflect_mcp/cli/ops/wake.py)",
      "category": "permissions",
      "severity": "high",
      "confidence": "strong",
      "description": "In the server's implementation (`reflect_mcp/cli/ops/wake.py:199`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
      "remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
      "location": {
        "kind": "server",
        "name": "reflect_mcp/cli/ops/wake.py"
      },
      "evidence": "pen(\"w\") proc = subprocess.Popen( [sys.executable, \"-m\", \"reflect_mcp.runtime.wake\"], en",
      "owasp": "LLM05:2025 Improper Output Handling",
      "data": {
        "rule": "MTC-SRC-002",
        "file": "reflect_mcp/cli/ops/wake.py",
        "line": 199,
        "nonRuntime": false
      }
    },
    {
      "ruleId": "MTC-SRC-002",
      "title": "Shell/command execution in server code (reflect_mcp/cli/track.py)",
      "category": "permissions",
      "severity": "high",
      "confidence": "strong",
      "description": "In the server's implementation (`reflect_mcp/cli/track.py:273`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
      "remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
      "location": {
        "kind": "server",
        "name": "reflect_mcp/cli/track.py"
      },
      "evidence": "_path] try: subprocess.run(cmd) except KeyboardInterrupt: pass def cmd_track_logs(args) -> Non",
      "owasp": "LLM05:2025 Improper Output Handling",
      "data": {
        "rule": "MTC-SRC-002",
        "file": "reflect_mcp/cli/track.py",
        "line": 273,
        "nonRuntime": false
      }
    },
    {
      "ruleId": "MTC-SRC-002",
      "title": "Shell/command execution in server code (reflect_mcp/cli/tui/dashboard_app.py)",
      "category": "permissions",
      "severity": "high",
      "confidence": "strong",
      "description": "In the server's implementation (`reflect_mcp/cli/tui/dashboard_app.py:1039`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
      "remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
      "location": {
        "kind": "server",
        "name": "reflect_mcp/cli/tui/dashboard_app.py"
      },
      "evidence": ".suspend(): subprocess.run([\"less\", \"-R\", tmp_path], check=False) import os as _os try:",
      "owasp": "LLM05:2025 Improper Output Handling",
      "data": {
        "rule": "MTC-SRC-002",
        "file": "reflect_mcp/cli/tui/dashboard_app.py",
        "line": 1039,
        "nonRuntime": false
      }
    },
    {
      "ruleId": "MTC-SRC-002",
      "title": "Shell/command execution in server code (reflect_mcp/cli/tui/inbox_app.py)",
      "category": "permissions",
      "severity": "high",
      "confidence": "strong",
      "description": "In the server's implementation (`reflect_mcp/cli/tui/inbox_app.py:693`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
      "remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
      "location": {
        "kind": "server",
        "name": "reflect_mcp/cli/tui/inbox_app.py"
      },
      "evidence": ".suspend(): subprocess.run([editor, str(path)]) # v0.4.91 (item 33 escalation): selection-mode toggle.",
      "owasp": "LLM05:2025 Improper Output Handling",
      "data": {
        "rule": "MTC-SRC-002",
        "file": "reflect_mcp/cli/tui/inbox_app.py",
        "line": 693,
        "nonRuntime": false
      }
    },
    {
      "ruleId": "MTC-SRC-002",
      "title": "Shell/command execution in server code (reflect_mcp/cli/tui/session_view.py)",
      "category": "permissions",
      "severity": "high",
      "confidence": "strong",
      "description": "In the server's implementation (`reflect_mcp/cli/tui/session_view.py:625`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
      "remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
      "location": {
        "kind": "server",
        "name": "reflect_mcp/cli/tui/session_view.py"
      },
      "evidence": "result = subprocess.run( [\"reflect\", \"wake\", \"start\", self._track],",
      "owasp": "LLM05:2025 Improper Output Handling",
      "data": {
        "rule": "MTC-SRC-002",
        "file": "reflect_mcp/cli/tui/session_view.py",
        "line": 625,
        "nonRuntime": false
      }
    },
    {
      "ruleId": "MTC-SRC-002",
      "title": "Shell/command execution in server code (reflect_mcp/cli/wake.py)",
      "category": "permissions",
      "severity": "high",
      "confidence": "strong",
      "description": "In the server's implementation (`reflect_mcp/cli/wake.py:200`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.",
      "remediation": "Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.",
      "location": {
        "kind": "server",
        "name": "reflect_mcp/cli/wake.py"
      },
      "evidence": "pen(\"w\") proc = subprocess.Popen( [sys.executable, \"-m\", \"reflect_mcp.runtime.wake\"], en",
      "owasp": "LLM05:2025 Improper Output Handling",
      "data": {
        "rule": "MTC-SRC-002",
        "file": "reflect_mcp/cli/wake.py",
        "line": 200,
        "nonRuntime": false
      }
    },
    {
      "ruleId": "MTC-SUP-012",
      "title": "Package has no license",
      "category": "hygiene",
      "severity": "info",
      "confidence": "strong",
      "description": "\"reflect-mcp\" has no declared license. This is a legal/reuse concern, not a security finding.",
      "location": {
        "kind": "package",
        "name": "reflect-mcp"
      }
    },
    {
      "ruleId": "MTC-SUP-011",
      "title": "Package declares no source repository",
      "category": "supply-chain",
      "severity": "info",
      "confidence": "strong",
      "description": "\"reflect-mcp\" declares no repository URL, so its published artifact cannot be compared against reviewable source.",
      "remediation": "Prefer packages that link to public, reviewable source.",
      "location": {
        "kind": "package",
        "name": "reflect-mcp"
      }
    }
  ],
  "toxicFlows": [],
  "capabilities": [
    {
      "tool": "track",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "track_admin",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "project",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "wake",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "cost",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "inbox",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "followup",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "schedule",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "docs",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "report",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "doctor",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "reflect_doctor",
      "tags": [],
      "reasons": {}
    }
  ],
  "surfaceDigest": "503bd346d11ab1da7e49346e7484127e81511417055b0243d3f9b7db73626eaf",
  "stats": {
    "tools": 12,
    "prompts": 0,
    "resources": 0,
    "findingsBySeverity": {
      "critical": 0,
      "high": 11,
      "medium": 0,
      "low": 0,
      "info": 2
    }
  }
}