{ "skill_name": "azure-ai-evaluation-py", "skill_path": "/workspace/skills/microsoft-agent-skills/.github/skills/azure-ai-evaluation-py", "skill_directory": "/workspace/skills/microsoft-agent-skills/.github/skills/azure-ai-evaluation-py", "is_safe": true, "max_severity": "MEDIUM", "findings_count": 2, "findings": [ { "id": "llm_finding_azure-ai-evaluation-py_0", "rule_id": "LLM_DATA_EXFILTRATION", "category": "data_exfiltration", "severity": "MEDIUM", "title": "Hardcoded API Key in Environment Variable Instructions", "description": "The skill instructions explicitly guide users to set AZURE_OPENAI_API_KEY as an environment variable and use it directly in code. While environment variables are better than hardcoding in source, the instructions don't emphasize secure credential management practices like Azure Key Vault or managed identities. The script does support DefaultAzureCredential as an alternative, but the primary examples use API keys.", "file_path": "SKILL.md", "line_number": null, "snippet": "```bash\nAZURE_OPENAI_API_KEY=\n```\nand\n```python\nmodel_config = {\n \"api_key\": os.environ[\"AZURE_OPENAI_API_KEY\"]\n}\n```", "remediation": "Prioritize DefaultAzureCredential in documentation and examples. Add warnings about API key security. Recommend Azure Key Vault or managed identities for production use. The script already supports credential-based auth, so update examples to make it the default pattern.", "analyzer": "llm", "metadata": { "model": "claude-sonnet-4-5-20250929", "overall_assessment": "The azure-ai-evaluation-py skill is a legitimate Azure SDK wrapper for evaluating AI applications. The skill itself is not malicious, but it has MEDIUM severity security concerns around credential management practices. The primary issues are: (1) instructions that emphasize API key usage over more secure credential methods, (2) connection string exposure risks without security guidance, and (3) missing package integrity verification. The skill correctly supports DefaultAzureCredential for secure authentication, but the documentation doesn't prioritize this approach. No prompt injection, data exfiltration, or malicious code patterns were detected. The referenced files are internal documentation and examples, which is normal for SDK skills. Overall, this is a safe skill that needs improved security guidance in its documentation.", "primary_threats": [ "Credential exposure via environment variables", "Missing security best practices in documentation", "Potential resource exhaustion in batch operations" ], "aitech": "AITech-8.2", "aitech_name": "Data Exfiltration / Exposure", "aisubtech": "AISubtech-8.2.3", "aisubtech_name": "Data Exfiltration via Agent Tooling", "scanner_category": "SECURITY VIOLATION", "meta_validated": true, "meta_confidence": "HIGH", "meta_confidence_reason": "LLM analyzer correctly identified that while the code supports secure authentication, the documentation doesn't emphasize it. This is a real documentation issue that could lead users to less secure practices.", "meta_exploitability": "Low - Requires user to follow insecure documentation patterns", "meta_impact": "Medium - Could lead to credential exposure if users don't use secure alternatives" } }, { "id": "llm_finding_azure-ai-evaluation-py_1", "rule_id": "LLM_DATA_EXFILTRATION", "category": "data_exfiltration", "severity": "MEDIUM", "title": "Connection String Exposure Risk", "description": "The skill requires AIPROJECT_CONNECTION_STRING environment variable for Foundry integration. Connection strings often contain sensitive credentials. The instructions don't provide guidance on secure storage or rotation of these connection strings.", "file_path": "SKILL.md", "line_number": 18, "snippet": "```bash\nAIPROJECT_CONNECTION_STRING=\n```", "remediation": "Add security guidance for connection string management. Recommend using Azure Key Vault references or managed identities instead of plain connection strings. Warn users not to commit connection strings to version control.", "analyzer": "llm", "metadata": { "model": "claude-sonnet-4-5-20250929", "overall_assessment": "The azure-ai-evaluation-py skill is a legitimate Azure SDK wrapper for evaluating AI applications. The skill itself is not malicious, but it has MEDIUM severity security concerns around credential management practices. The primary issues are: (1) instructions that emphasize API key usage over more secure credential methods, (2) connection string exposure risks without security guidance, and (3) missing package integrity verification. The skill correctly supports DefaultAzureCredential for secure authentication, but the documentation doesn't prioritize this approach. No prompt injection, data exfiltration, or malicious code patterns were detected. The referenced files are internal documentation and examples, which is normal for SDK skills. Overall, this is a safe skill that needs improved security guidance in its documentation.", "primary_threats": [ "Credential exposure via environment variables", "Missing security best practices in documentation", "Potential resource exhaustion in batch operations" ], "aitech": "AITech-8.2", "aitech_name": "Data Exfiltration / Exposure", "aisubtech": "AISubtech-8.2.3", "aisubtech_name": "Data Exfiltration via Agent Tooling", "scanner_category": "SECURITY VIOLATION", "meta_validated": true, "meta_confidence": "HIGH", "meta_confidence_reason": "Connection strings are sensitive credentials. Missing security guidance is a real documentation gap that could lead to credential exposure.", "meta_exploitability": "Low - Requires user to mishandle credentials", "meta_impact": "Medium - Connection string exposure could allow unauthorized access to Foundry projects" } } ], "scan_duration_seconds": 30.666600704193115, "duration_ms": 30666, "analyzers_used": [ "static_analyzer", "behavioral_analyzer", "llm_analyzer", "trigger_analyzer", "meta_analyzer" ], "timestamp": "2026-02-05T13:26:55.390153" }