Raw scan report

Agent Memory — 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": "@xultrax-web/agent-memory-mcp",
    "source": {
      "kind": "package",
      "origin": "@xultrax-web/agent-memory-mcp"
    },
    "server": {
      "name": "@xultrax-web/agent-memory-mcp"
    }
  },
  "grade": "C",
  "score": {
    "score": 77,
    "threatScore": 84,
    "grade": "C",
    "band": "C",
    "categorySubtotals": {
      "injection": 16,
      "exfiltration": 0,
      "permissions": 0,
      "supply-chain": 0,
      "network": 0,
      "hygiene": 0
    },
    "vector": [
      {
        "kind": "threat",
        "ruleId": "MTC-INJ-CMD-1",
        "category": "injection",
        "severity": "high",
        "confidence": "strong",
        "rawWeight": 22,
        "confidenceMult": 0.7,
        "diminishingFactor": 1,
        "appliedPenalty": 15.4
      },
      {
        "kind": "threat",
        "ruleId": "MTC-INJ-CMD-1",
        "category": "injection",
        "severity": "low",
        "confidence": "heuristic",
        "rawWeight": 3,
        "confidenceMult": 0.4,
        "diminishingFactor": 0.5,
        "appliedPenalty": 0.6
      },
      {
        "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": "repo",
        "label": "publisher verification (public source) — no provenance, but the source is public and inspectable",
        "appliedPenalty": 1
      },
      {
        "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": [
      "ingests untrusted external content (a prompt-injection entry point)",
      "can send data / act on an external service",
      "can execute shell commands or code"
    ],
    "tags": [
      "untrusted-input",
      "external-sink",
      "code-exec"
    ]
  },
  "coverage": {
    "level": "source",
    "inputs": {
      "toolSurface": true,
      "implementationSource": true,
      "packageMetadata": true,
      "liveTransport": false
    },
    "caveats": [
      "Tools were statically extracted from the published source (36 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-INJ-CMD-1",
      "title": "Shell command embedded in tool metadata",
      "category": "injection",
      "severity": "high",
      "confidence": "strong",
      "description": "Shell command embedded in tool metadata detected in the description of tool \"init\". Instruction-like content in tool metadata is executed by the model, not the human, and is the primary tool-poisoning vector.",
      "remediation": "Tool descriptions should describe behavior, not instruct the assistant. Treat imperative / secrecy / sequencing language in metadata as hostile.",
      "location": {
        "kind": "tool",
        "name": "init",
        "field": "description"
      },
      "evidence": "rm -rf",
      "owasp": "LLM01:2025 Prompt Injection",
      "data": {
        "kind": "command-in-prose"
      }
    },
    {
      "ruleId": "MTC-CAP-001",
      "title": "Tool \"no-curl-pipe-shell\" exposes command/code execution",
      "category": "permissions",
      "severity": "high",
      "confidence": "strong",
      "description": "Tool \"no-curl-pipe-shell\" appears to run shell commands or evaluate code (keyword \"shell\" in tool name). Arbitrary execution driven by model input is one of the most dangerous MCP capabilities; combined with any untrusted input it becomes RCE.",
      "remediation": "Sandbox execution, allowlist commands/arguments, and never pass model output to a shell unescaped.",
      "location": {
        "kind": "tool",
        "name": "no-curl-pipe-shell"
      },
      "owasp": "LLM05:2025 Improper Output Handling",
      "data": {
        "tags": [
          "external-sink",
          "code-exec"
        ]
      }
    },
    {
      "ruleId": "MTC-SRC-002",
      "title": "Shell/command execution in server code (dist/index.js)",
      "category": "permissions",
      "severity": "high",
      "confidence": "strong",
      "description": "In the server's implementation (`dist/index.js:29`): 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": "dist/index.js"
      },
      "evidence": "spawnSync } from \"node:child_process\"; import { createHash, createHmac, createPrivateKey, createPublicKey, generateKeyP",
      "owasp": "LLM05:2025 Improper Output Handling",
      "data": {
        "rule": "MTC-SRC-002",
        "file": "dist/index.js",
        "line": 29,
        "nonRuntime": false
      }
    },
    {
      "ruleId": "MTC-SRC-002",
      "title": "Shell/command execution in server code (dist/tui.js)",
      "category": "permissions",
      "severity": "high",
      "confidence": "strong",
      "description": "In the server's implementation (`dist/tui.js:26`): 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": "dist/tui.js"
      },
      "evidence": "spawnSync } from \"node:child_process\"; import { useEffect, useMemo, useState } from \"react\"; import Fuse from \"fuse.js\"",
      "owasp": "LLM05:2025 Improper Output Handling",
      "data": {
        "rule": "MTC-SRC-002",
        "file": "dist/tui.js",
        "line": 26,
        "nonRuntime": false
      }
    },
    {
      "ruleId": "MTC-FLOW-005",
      "title": "Untrusted input can drive an external action",
      "category": "exfiltration",
      "severity": "medium",
      "confidence": "strong",
      "description": "Untrusted-input tools ([get_memory]) co-exist with external-action tools ([no-curl-pipe-shell]). A prompt injection could cause unwanted external actions, though no direct sensitive-data leak path was found.",
      "remediation": "Require confirmation for state-changing/egress actions triggered after processing untrusted content.",
      "location": {
        "kind": "flow",
        "name": "get_memory → no-curl-pipe-shell"
      },
      "evidence": "untrusted [get_memory] → sinks [no-curl-pipe-shell]",
      "owasp": "LLM06:2025 Excessive Agency",
      "references": [],
      "data": {
        "untrusted": [
          "get_memory"
        ],
        "sinks": [
          "no-curl-pipe-shell"
        ]
      }
    },
    {
      "ruleId": "MTC-CAP-005",
      "title": "Mutating tool \"no-curl-pipe-shell\" declares no destructiveHint",
      "category": "hygiene",
      "severity": "low",
      "confidence": "heuristic",
      "description": "Tool \"no-curl-pipe-shell\" can mutate/egress but declares no destructiveHint. Clients that don't default to spec-safe behavior may not prompt before running it.",
      "remediation": "Declare accurate annotations, and gate destructive tools on user confirmation regardless.",
      "location": {
        "kind": "tool",
        "name": "no-curl-pipe-shell"
      },
      "data": {
        "tags": [
          "external-sink",
          "code-exec"
        ]
      }
    },
    {
      "ruleId": "MTC-INJ-CMD-1",
      "title": "Shell command embedded in tool metadata",
      "category": "injection",
      "severity": "low",
      "confidence": "heuristic",
      "description": "Shell command embedded in tool metadata detected in the description of tool \"no-rm-rf\". Instruction-like content in tool metadata is executed by the model, not the human, and is the primary tool-poisoning vector.",
      "remediation": "Tool descriptions should describe behavior, not instruct the assistant. Treat imperative / secrecy / sequencing language in metadata as hostile.",
      "location": {
        "kind": "tool",
        "name": "no-rm-rf",
        "field": "description"
      },
      "evidence": "rm -rf",
      "owasp": "LLM01:2025 Prompt Injection",
      "data": {
        "kind": "command-in-prose"
      }
    }
  ],
  "toxicFlows": [
    {
      "id": "flow-1",
      "severity": "medium",
      "confidence": "strong",
      "untrustedInput": [
        "get_memory"
      ],
      "sensitiveSource": [],
      "externalSink": [
        "no-curl-pipe-shell"
      ],
      "selfContained": false,
      "description": "Untrusted input can drive an external action even though no sensitive source is exposed."
    }
  ],
  "capabilities": [
    {
      "tool": "Memory index",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "extract_memories",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "summarize_topic",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "topic",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "prepare_handoff",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "project",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "audit_stale",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "save_memory",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "search_memories",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "relevant_memories",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "get_memory",
      "tags": [
        "untrusted-input"
      ],
      "reasons": {
        "untrusted-input": [
          "keyword \"fetch\""
        ]
      }
    },
    {
      "tool": "list_memories",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "delete_memory",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "restore_memory",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "doctor",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "stats",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "rotate_key",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "init",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "validate_receipt",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "log_events",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "verify_memory",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "find_backlinks",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "find_related",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "sync_status",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "sync_push",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "sync_pull",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "save_rule",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "list_rules",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "audit",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "check_action",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "emit_companions",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "protect-main-branch",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "no-rm-rf",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "no-prod-data-destruction",
      "tags": [],
      "reasons": {}
    },
    {
      "tool": "no-curl-pipe-shell",
      "tags": [
        "external-sink",
        "code-exec"
      ],
      "reasons": {
        "external-sink": [
          "keyword \"curl\""
        ],
        "code-exec": [
          "keyword \"shell\" in tool name"
        ]
      }
    },
    {
      "tool": "no-secrets-in-commits",
      "tags": [],
      "reasons": {}
    }
  ],
  "surfaceDigest": "6507a7fa7d8269859a1f39fec643b62c8da9d8a97d696f29a324f470526732fe",
  "stats": {
    "tools": 36,
    "prompts": 0,
    "resources": 0,
    "findingsBySeverity": {
      "critical": 0,
      "high": 4,
      "medium": 1,
      "low": 2,
      "info": 0
    }
  }
}