Agentdb — the complete, unedited output of the deterministic mcptrustchecker engine v1.5.0, scanned . Every finding, capability tag and score component below is exactly what the engine produced — no AI, no post-processing.
{
"tool": {
"name": "mcptrustchecker",
"version": "1.5.0",
"methodologyVersion": "mcptrustchecker-1.4"
},
"target": {
"id": "agentdb",
"source": {
"kind": "package",
"origin": "agentdb"
},
"server": {
"name": "agentdb"
}
},
"grade": "B",
"score": {
"score": 86,
"grade": "B",
"band": "B",
"categorySubtotals": {
"injection": 0,
"exfiltration": 0,
"permissions": 13.55,
"supply-chain": 0,
"network": 0,
"hygiene": 0
},
"vector": [
{
"ruleId": "MTC-SRC-007",
"category": "permissions",
"severity": "medium",
"confidence": "strong",
"rawWeight": 9,
"confidenceMult": 0.7,
"diminishingFactor": 1,
"appliedPenalty": 6.3
},
{
"ruleId": "MTC-SRC-007",
"category": "permissions",
"severity": "medium",
"confidence": "strong",
"rawWeight": 9,
"confidenceMult": 0.7,
"diminishingFactor": 0.5,
"appliedPenalty": 3.15
},
{
"ruleId": "MTC-SRC-007",
"category": "permissions",
"severity": "medium",
"confidence": "strong",
"rawWeight": 9,
"confidenceMult": 0.7,
"diminishingFactor": 0.25,
"appliedPenalty": 1.58
},
{
"ruleId": "MTC-SRC-007",
"category": "permissions",
"severity": "medium",
"confidence": "strong",
"rawWeight": 9,
"confidenceMult": 0.7,
"diminishingFactor": 0.1,
"appliedPenalty": 0.63
},
{
"ruleId": "MTC-SRC-007",
"category": "permissions",
"severity": "medium",
"confidence": "strong",
"rawWeight": 9,
"confidenceMult": 0.7,
"diminishingFactor": 0.1,
"appliedPenalty": 0.63
},
{
"ruleId": "MTC-SRC-007",
"category": "permissions",
"severity": "medium",
"confidence": "strong",
"rawWeight": 9,
"confidenceMult": 0.7,
"diminishingFactor": 0.1,
"appliedPenalty": 0.63
},
{
"ruleId": "MTC-SRC-007",
"category": "permissions",
"severity": "medium",
"confidence": "strong",
"rawWeight": 9,
"confidenceMult": 0.7,
"diminishingFactor": 0.1,
"appliedPenalty": 0.63
}
],
"gatesFired": [],
"methodologyVersion": "mcptrustchecker-1.4"
},
"capability": {
"level": "high",
"reasons": [
"can send data / act on an external service",
"can execute shell commands or code"
],
"tags": [
"code-exec",
"external-sink"
]
},
"coverage": {
"level": "source",
"inputs": {
"toolSurface": false,
"implementationSource": true,
"packageMetadata": true,
"liveTransport": false
},
"caveats": [
"No tools were enumerated, so prompt-injection, capability and toxic-flow analysis had no tool surface to inspect. To grade a package’s real runtime tools, scan the running server: --command \"npx -y <package>\"."
]
},
"findings": [
{
"ruleId": "MTC-SRC-001",
"title": "Dynamic code execution in server code (openagent/compiler/resources/main.js)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`openagent/compiler/resources/main.js:1`): 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": "openagent/compiler/resources/main.js"
},
"evidence": "Function(\"",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-001",
"file": "openagent/compiler/resources/main.js",
"line": 1
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (openagent/agent/agentbox/tinybox.py)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`openagent/agent/agentbox/tinybox.py:93`): 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": "openagent/agent/agentbox/tinybox.py"
},
"evidence": "subprocess.Popen(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "openagent/agent/agentbox/tinybox.py",
"line": 93
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (openagent/compiler/library/_shell.py)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`openagent/compiler/library/_shell.py:37`): 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": "openagent/compiler/library/_shell.py"
},
"evidence": "subprocess.Popen(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "openagent/compiler/library/_shell.py",
"line": 37
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (openagent/fact_check/code/helper/_execution.py)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`openagent/fact_check/code/helper/_execution.py:83`): 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": "openagent/fact_check/code/helper/_execution.py"
},
"evidence": "exec(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "openagent/fact_check/code/helper/_execution.py",
"line": 83
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (openagent/fact_check/math/tool.py)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`openagent/fact_check/math/tool.py:15`): 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": "openagent/fact_check/math/tool.py"
},
"evidence": "exec(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "openagent/fact_check/math/tool.py",
"line": 15
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (openagent/helpers/bash.py)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`openagent/helpers/bash.py:78`): 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": "openagent/helpers/bash.py"
},
"evidence": "subprocess.run(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "openagent/helpers/bash.py",
"line": 78
}
},
{
"ruleId": "MTC-SRC-002",
"title": "Shell/command execution in server code (openagent/helpers/math.py)",
"category": "permissions",
"severity": "high",
"confidence": "strong",
"description": "In the server's implementation (`openagent/helpers/math.py: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": "openagent/helpers/math.py"
},
"evidence": "exec(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-002",
"file": "openagent/helpers/math.py",
"line": 19
}
},
{
"ruleId": "MTC-SRC-003",
"title": "Hardcoded egress to an external endpoint (openagent/rag/document_loaders/dad_jokes/base.py)",
"category": "exfiltration",
"severity": "medium",
"confidence": "strong",
"description": "In the server's implementation (`openagent/rag/document_loaders/dad_jokes/base.py:18`): A hardcoded outbound call to a fixed external host inside server code is a classic exfiltration/telemetry channel — especially paired with reads of local data. 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": "openagent/rag/document_loaders/dad_jokes/base.py"
},
"evidence": "requests.get(\n \"https://icanhazdadjoke.com/\"",
"owasp": "LLM02:2025 Sensitive Information Disclosure",
"data": {
"rule": "MTC-SRC-003",
"file": "openagent/rag/document_loaders/dad_jokes/base.py",
"line": 18
}
},
{
"ruleId": "MTC-SRC-003",
"title": "Hardcoded egress to an external endpoint (openagent/rag/document_loaders/file/audio_gladia/base.py)",
"category": "exfiltration",
"severity": "medium",
"confidence": "strong",
"description": "In the server's implementation (`openagent/rag/document_loaders/file/audio_gladia/base.py:89`): A hardcoded outbound call to a fixed external host inside server code is a classic exfiltration/telemetry channel — especially paired with reads of local data. 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": "openagent/rag/document_loaders/file/audio_gladia/base.py"
},
"evidence": "requests.post(\n \"https://api.gladia.io/audio/text/audio-transcription/\"",
"owasp": "LLM02:2025 Sensitive Information Disclosure",
"data": {
"rule": "MTC-SRC-003",
"file": "openagent/rag/document_loaders/file/audio_gladia/base.py",
"line": 89
}
},
{
"ruleId": "MTC-SRC-003",
"title": "Hardcoded egress to an external endpoint (openagent/rag/document_loaders/papers/pubmed/base.py)",
"category": "exfiltration",
"severity": "medium",
"confidence": "strong",
"description": "In the server's implementation (`openagent/rag/document_loaders/papers/pubmed/base.py:38`): A hardcoded outbound call to a fixed external host inside server code is a classic exfiltration/telemetry channel — especially paired with reads of local data. 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": "openagent/rag/document_loaders/papers/pubmed/base.py"
},
"evidence": "requests.get(\n \"https://eutils.ncbi.nlm.nih.gov/entrez/eutils/esearch.fcgi\"",
"owasp": "LLM02:2025 Sensitive Information Disclosure",
"data": {
"rule": "MTC-SRC-003",
"file": "openagent/rag/document_loaders/papers/pubmed/base.py",
"line": 38
}
},
{
"ruleId": "MTC-SRC-007",
"title": "Unsafe deserialization (openagent/fact_check/code/pipeline.py)",
"category": "permissions",
"severity": "medium",
"confidence": "strong",
"description": "In the server's implementation (`openagent/fact_check/code/pipeline.py:25`): Deserializing untrusted data with these APIs can execute arbitrary code (a well-known RCE gadget class). 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": "openagent/fact_check/code/pipeline.py"
},
"evidence": "yaml.load(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-007",
"file": "openagent/fact_check/code/pipeline.py",
"line": 25
}
},
{
"ruleId": "MTC-SRC-007",
"title": "Unsafe deserialization (openagent/fact_check/knowledge_qa/pipeline.py)",
"category": "permissions",
"severity": "medium",
"confidence": "strong",
"description": "In the server's implementation (`openagent/fact_check/knowledge_qa/pipeline.py:21`): Deserializing untrusted data with these APIs can execute arbitrary code (a well-known RCE gadget class). 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": "openagent/fact_check/knowledge_qa/pipeline.py"
},
"evidence": "yaml.load(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-007",
"file": "openagent/fact_check/knowledge_qa/pipeline.py",
"line": 21
}
},
{
"ruleId": "MTC-SRC-007",
"title": "Unsafe deserialization (openagent/fact_check/math/pipeline.py)",
"category": "permissions",
"severity": "medium",
"confidence": "strong",
"description": "In the server's implementation (`openagent/fact_check/math/pipeline.py:18`): Deserializing untrusted data with these APIs can execute arbitrary code (a well-known RCE gadget class). 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": "openagent/fact_check/math/pipeline.py"
},
"evidence": "yaml.load(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-007",
"file": "openagent/fact_check/math/pipeline.py",
"line": 18
}
},
{
"ruleId": "MTC-SRC-007",
"title": "Unsafe deserialization (openagent/fact_check/med_doc_qa/pipeline.py)",
"category": "permissions",
"severity": "medium",
"confidence": "strong",
"description": "In the server's implementation (`openagent/fact_check/med_doc_qa/pipeline.py:14`): Deserializing untrusted data with these APIs can execute arbitrary code (a well-known RCE gadget class). 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": "openagent/fact_check/med_doc_qa/pipeline.py"
},
"evidence": "yaml.load(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-007",
"file": "openagent/fact_check/med_doc_qa/pipeline.py",
"line": 14
}
},
{
"ruleId": "MTC-SRC-007",
"title": "Unsafe deserialization (openagent/fact_check/scientific/pipeline.py)",
"category": "permissions",
"severity": "medium",
"confidence": "strong",
"description": "In the server's implementation (`openagent/fact_check/scientific/pipeline.py:18`): Deserializing untrusted data with these APIs can execute arbitrary code (a well-known RCE gadget class). 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": "openagent/fact_check/scientific/pipeline.py"
},
"evidence": "yaml.load(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-007",
"file": "openagent/fact_check/scientific/pipeline.py",
"line": 18
}
},
{
"ruleId": "MTC-SRC-007",
"title": "Unsafe deserialization (openagent/fact_check/utils/base/pipeline.py)",
"category": "permissions",
"severity": "medium",
"confidence": "strong",
"description": "In the server's implementation (`openagent/fact_check/utils/base/pipeline.py:15`): Deserializing untrusted data with these APIs can execute arbitrary code (a well-known RCE gadget class). 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": "openagent/fact_check/utils/base/pipeline.py"
},
"evidence": "yaml.load(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-007",
"file": "openagent/fact_check/utils/base/pipeline.py",
"line": 15
}
},
{
"ruleId": "MTC-SRC-007",
"title": "Unsafe deserialization (openagent/fact_check/utils/claim_extractor.py)",
"category": "permissions",
"severity": "medium",
"confidence": "strong",
"description": "In the server's implementation (`openagent/fact_check/utils/claim_extractor.py:58`): Deserializing untrusted data with these APIs can execute arbitrary code (a well-known RCE gadget class). 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": "openagent/fact_check/utils/claim_extractor.py"
},
"evidence": "yaml.load(",
"owasp": "LLM05:2025 Improper Output Handling",
"data": {
"rule": "MTC-SRC-007",
"file": "openagent/fact_check/utils/claim_extractor.py",
"line": 58
}
},
{
"ruleId": "MTC-SUP-012",
"title": "Package has no license",
"category": "hygiene",
"severity": "info",
"confidence": "strong",
"description": "\"agentdb\" has no declared license. This is a legal/reuse concern, not a security finding.",
"location": {
"kind": "package",
"name": "agentdb"
}
}
],
"toxicFlows": [],
"capabilities": [],
"surfaceDigest": "e020bd1c26f92c5f70125933078fb00933d7e1abb0ba7b3b60dc972c38665f9e",
"stats": {
"tools": 0,
"prompts": 0,
"resources": 0,
"findingsBySeverity": {
"critical": 0,
"high": 7,
"medium": 10,
"low": 0,
"info": 1
}
}
}