Vouch Kb — the complete, unedited output of the deterministic mcptrustchecker engine v1.7.0, scanned . Every finding, capability tag and score component below is exactly what the engine produced — no AI, no post-processing.
{
"tool": {
"name": "mcptrustchecker",
"version": "1.7.0",
"methodologyVersion": "mcptrustchecker-1.7"
},
"target": {
"id": "vouch-kb",
"source": {
"kind": "package",
"origin": "vouch-kb"
},
"server": {
"name": "vouch-kb"
}
},
"grade": "D",
"score": {
"score": 60,
"threatScore": 67,
"grade": "D",
"band": "D",
"gateCap": "D",
"categorySubtotals": {
"injection": 0,
"exfiltration": 33,
"permissions": 0,
"supply-chain": 0,
"network": 0,
"hygiene": 0
},
"vector": [
{
"kind": "threat",
"ruleId": "MTC-SRC-008",
"category": "exfiltration",
"severity": "high",
"confidence": "confirmed",
"rawWeight": 22,
"confidenceMult": 1,
"diminishingFactor": 1,
"appliedPenalty": 22
},
{
"kind": "threat",
"ruleId": "MTC-SRC-008",
"category": "exfiltration",
"severity": "high",
"confidence": "confirmed",
"rawWeight": 22,
"confidenceMult": 1,
"diminishingFactor": 0.5,
"appliedPenalty": 11
},
{
"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": [
"2 confirmed high findings → grade capped at D"
],
"methodologyVersion": "mcptrustchecker-1.7"
},
"capability": {
"level": "high",
"reasons": [
"can send data / act on an external service",
"reads sensitive or local data",
"has a read → egress path (a data-exfiltration surface)",
"can execute shell commands or code"
],
"tags": [
"external-sink",
"code-exec",
"sensitive-source"
]
},
"coverage": {
"level": "source",
"inputs": {
"toolSurface": true,
"implementationSource": true,
"packageMetadata": true,
"liveTransport": false
},
"caveats": [
"Tools were statically extracted from the published source (68 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-006",
"title": "Reads a sensitive credential path or dumps the environment (tests/test_install_adapter.py)",
"category": "exfiltration",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`tests/test_install_adapter.py:1118`): Reading private keys / cloud credentials, or serializing the whole environment, is a sensitive-data source that becomes exfiltration when combined with any egress. 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": "tests/test_install_adapter.py"
},
"evidence": "dict(os.environ)",
"owasp": "LLM02:2025 Sensitive Information Disclosure",
"data": {
"rule": "MTC-SRC-006",
"file": "tests/test_install_adapter.py",
"line": 1118
}
},
{
"ruleId": "MTC-SRC-006",
"title": "Reads a sensitive credential path or dumps the environment (tests/test_jsonl_server.py)",
"category": "exfiltration",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`tests/test_jsonl_server.py:114`): Reading private keys / cloud credentials, or serializing the whole environment, is a sensitive-data source that becomes exfiltration when combined with any egress. 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": "tests/test_jsonl_server.py"
},
"evidence": "~/.ssh",
"owasp": "LLM02:2025 Sensitive Information Disclosure",
"data": {
"rule": "MTC-SRC-006",
"file": "tests/test_jsonl_server.py",
"line": 114
}
},
{
"ruleId": "MTC-SRC-008",
"title": "Hardcoded AWS access key id in server code (tests/test_capture.py)",
"category": "exfiltration",
"severity": "high",
"confidence": "confirmed",
"description": "A live-looking AWS access key id is hardcoded in `tests/test_capture.py:78`. Secrets in source ship to everyone who installs the package and are a direct credential leak.",
"remediation": "Remove the secret, rotate it, and load credentials from the environment or a secret store.",
"location": {
"kind": "server",
"name": "tests/test_capture.py"
},
"evidence": "AWS access key id: AKIA…(redacted)",
"owasp": "LLM02:2025 Sensitive Information Disclosure",
"data": {
"secretType": "aws-access-key",
"file": "tests/test_capture.py",
"line": 78
}
},
{
"ruleId": "MTC-SRC-008",
"title": "Hardcoded AWS access key id in server code (tests/test_secrets.py)",
"category": "exfiltration",
"severity": "high",
"confidence": "confirmed",
"description": "A live-looking AWS access key id is hardcoded in `tests/test_secrets.py:16`. Secrets in source ship to everyone who installs the package and are a direct credential leak.",
"remediation": "Remove the secret, rotate it, and load credentials from the environment or a secret store.",
"location": {
"kind": "server",
"name": "tests/test_secrets.py"
},
"evidence": "AWS access key id: AKIA…(redacted)",
"owasp": "LLM02:2025 Sensitive Information Disclosure",
"data": {
"secretType": "aws-access-key",
"file": "tests/test_secrets.py",
"line": 16
}
},
{
"ruleId": "MTC-CAP-001",
"title": "Tool \"kb_eval_embeddings\" exposes command/code execution",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "Tool \"kb_eval_embeddings\" appears to run shell commands or evaluate code (keyword \"eval\"). 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": "kb_eval_embeddings"
},
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"tags": [
"code-exec"
]
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (adapters/openclaw/vouch-context-engine.mjs)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`adapters/openclaw/vouch-context-engine.mjs:19`): 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": "adapters/openclaw/vouch-context-engine.mjs"
},
"evidence": "child_process",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "adapters/openclaw/vouch-context-engine.mjs",
"line": 19
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (src/vouch/auto_pr.py)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`src/vouch/auto_pr.py:100`): 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": "src/vouch/auto_pr.py"
},
"evidence": "subprocess.run(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "src/vouch/auto_pr.py",
"line": 100
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (src/vouch/capture.py)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`src/vouch/capture.py:198`): 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": "src/vouch/capture.py"
},
"evidence": "subprocess.run(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "src/vouch/capture.py",
"line": 198
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (src/vouch/llm_draft.py)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`src/vouch/llm_draft.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": "src/vouch/llm_draft.py"
},
"evidence": "subprocess.run(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "src/vouch/llm_draft.py",
"line": 45
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (src/vouch/pr_cache.py)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`src/vouch/pr_cache.py:150`): 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": "src/vouch/pr_cache.py"
},
"evidence": "subprocess.run(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "src/vouch/pr_cache.py",
"line": 150
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (tests/test_capture.py)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`tests/test_capture.py:242`): 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": "tests/test_capture.py"
},
"evidence": "subprocess.run(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "tests/test_capture.py",
"line": 242
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (tests/test_cli.py)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`tests/test_cli.py:691`): 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": "tests/test_cli.py"
},
"evidence": "subprocess.run(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "tests/test_cli.py",
"line": 691
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (tests/test_codex_adapter_load_real.py)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`tests/test_codex_adapter_load_real.py:44`): 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": "tests/test_codex_adapter_load_real.py"
},
"evidence": "subprocess.run(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "tests/test_codex_adapter_load_real.py",
"line": 44
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (tests/test_hub.py)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`tests/test_hub.py:67`): 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": "tests/test_hub.py"
},
"evidence": "subprocess.run(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "tests/test_hub.py",
"line": 67
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (tests/test_install_adapter.py)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`tests/test_install_adapter.py:1080`): 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": "tests/test_install_adapter.py"
},
"evidence": "subprocess.run(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "tests/test_install_adapter.py",
"line": 1080
}
},
{
"ruleId": "MTC-SRC-005",
"title": "Dynamic module load from a non-literal (src/vouch/web/__init__.py)",
"category": "permissions",
"severity": "medium",
"confidence": "heuristic",
"description": "In the server's implementation (`src/vouch/web/__init__.py:44`): Loading a module chosen at runtime (from a variable) can pull in and run attacker-influenced code paths. 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": "src/vouch/web/__init__.py"
},
"evidence": "__import__(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-005",
"file": "src/vouch/web/__init__.py",
"line": 44
}
},
{
"ruleId": "MTC-CAP-005",
"title": "Mutating tool \"kb_eval_embeddings\" declares no destructiveHint",
"category": "hygiene",
"severity": "low",
"confidence": "heuristic",
"description": "Tool \"kb_eval_embeddings\" 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": "kb_eval_embeddings"
},
"data": {
"tags": [
"code-exec"
]
}
}
],
"toxicFlows": [],
"capabilities": [
{
"tool": "kb_capabilities",
"tags": [],
"reasons": {}
},
{
"tool": "kb_status",
"tags": [],
"reasons": {}
},
{
"tool": "kb_stats",
"tags": [],
"reasons": {}
},
{
"tool": "kb_list_skills",
"tags": [
"external-sink"
],
"reasons": {
"external-sink": [
"keyword \"publish\""
]
}
},
{
"tool": "kb_get_skill",
"tags": [
"external-sink"
],
"reasons": {
"external-sink": [
"keyword \"publish\""
]
}
},
{
"tool": "kb_activity",
"tags": [],
"reasons": {}
},
{
"tool": "kb_digest",
"tags": [],
"reasons": {}
},
{
"tool": "kb_search",
"tags": [],
"reasons": {}
},
{
"tool": "kb_experts",
"tags": [],
"reasons": {}
},
{
"tool": "kb_neighbors",
"tags": [],
"reasons": {}
},
{
"tool": "kb_context",
"tags": [],
"reasons": {}
},
{
"tool": "kb_synthesize",
"tags": [],
"reasons": {}
},
{
"tool": "kb_read_page",
"tags": [],
"reasons": {}
},
{
"tool": "kb_read_claim",
"tags": [],
"reasons": {}
},
{
"tool": "kb_read_entity",
"tags": [],
"reasons": {}
},
{
"tool": "kb_read_relation",
"tags": [],
"reasons": {}
},
{
"tool": "kb_diff",
"tags": [],
"reasons": {}
},
{
"tool": "kb_list_pages",
"tags": [],
"reasons": {}
},
{
"tool": "kb_list_claims",
"tags": [],
"reasons": {}
},
{
"tool": "kb_list_entities",
"tags": [],
"reasons": {}
},
{
"tool": "kb_list_relations",
"tags": [],
"reasons": {}
},
{
"tool": "kb_list_sources",
"tags": [],
"reasons": {}
},
{
"tool": "kb_list_pending",
"tags": [],
"reasons": {}
},
{
"tool": "kb_triage_pending",
"tags": [],
"reasons": {}
},
{
"tool": "kb_register_source",
"tags": [],
"reasons": {}
},
{
"tool": "kb_register_source_from_path",
"tags": [],
"reasons": {}
},
{
"tool": "kb_propose_claim",
"tags": [],
"reasons": {}
},
{
"tool": "kb_propose_page",
"tags": [],
"reasons": {}
},
{
"tool": "kb_compile",
"tags": [],
"reasons": {}
},
{
"tool": "kb_summarize_session",
"tags": [],
"reasons": {}
},
{
"tool": "kb_list_sessions",
"tags": [],
"reasons": {}
},
{
"tool": "kb_session_transcript",
"tags": [],
"reasons": {}
},
{
"tool": "kb_propose_entity",
"tags": [],
"reasons": {}
},
{
"tool": "kb_propose_relation",
"tags": [],
"reasons": {}
},
{
"tool": "kb_propose_delete",
"tags": [],
"reasons": {}
},
{
"tool": "kb_approve",
"tags": [],
"reasons": {}
},
{
"tool": "kb_reject",
"tags": [],
"reasons": {}
},
{
"tool": "kb_reject_extracted",
"tags": [],
"reasons": {}
},
{
"tool": "kb_expire",
"tags": [],
"reasons": {}
},
{
"tool": "kb_supersede",
"tags": [],
"reasons": {}
},
{
"tool": "kb_contradict",
"tags": [],
"reasons": {}
},
{
"tool": "kb_archive",
"tags": [],
"reasons": {}
},
{
"tool": "kb_confirm",
"tags": [],
"reasons": {}
},
{
"tool": "kb_clear_claims",
"tags": [],
"reasons": {}
},
{
"tool": "kb_cite",
"tags": [],
"reasons": {}
},
{
"tool": "kb_source_verify",
"tags": [],
"reasons": {}
},
{
"tool": "kb_session_start",
"tags": [],
"reasons": {}
},
{
"tool": "kb_volunteer_context",
"tags": [],
"reasons": {}
},
{
"tool": "kb_session_end",
"tags": [],
"reasons": {}
},
{
"tool": "kb_crystallize",
"tags": [],
"reasons": {}
},
{
"tool": "kb_index_rebuild",
"tags": [],
"reasons": {}
},
{
"tool": "kb_lint",
"tags": [],
"reasons": {}
},
{
"tool": "kb_doctor",
"tags": [],
"reasons": {}
},
{
"tool": "kb_export",
"tags": [],
"reasons": {}
},
{
"tool": "kb_export_check",
"tags": [],
"reasons": {}
},
{
"tool": "kb_import_check",
"tags": [],
"reasons": {}
},
{
"tool": "kb_audit",
"tags": [],
"reasons": {}
},
{
"tool": "kb_reindex_embeddings",
"tags": [],
"reasons": {}
},
{
"tool": "kb_dedup_scan",
"tags": [],
"reasons": {}
},
{
"tool": "kb_eval_embeddings",
"tags": [
"code-exec"
],
"reasons": {
"code-exec": [
"keyword \"eval\""
]
}
},
{
"tool": "kb_embeddings_stats",
"tags": [],
"reasons": {}
},
{
"tool": "kb_why",
"tags": [],
"reasons": {}
},
{
"tool": "kb_trace",
"tags": [],
"reasons": {}
},
{
"tool": "kb_impact",
"tags": [
"external-sink"
],
"reasons": {
"external-sink": [
"keyword \"forward\""
]
}
},
{
"tool": "kb_graph_export",
"tags": [],
"reasons": {}
},
{
"tool": "kb_provenance_rebuild",
"tags": [],
"reasons": {}
},
{
"tool": "kb_detect_themes",
"tags": [],
"reasons": {}
},
{
"tool": "kb_propose_theme",
"tags": [],
"reasons": {}
}
],
"surfaceDigest": "a355ab74e88a8369dae69d7b9a23ea42718b8499d95dcfe750ac20f8f3f3a4e7",
"stats": {
"tools": 68,
"prompts": 0,
"resources": 0,
"findingsBySeverity": {
"critical": 0,
"high": 15,
"medium": 1,
"low": 1,
"info": 0
}
}
}