Security risk report for @forgecat/microsoft-azure_skills-azure-compliance v0.0.5
Source Integrity
Low
Profile metadata attributes Microsoft as author with reference to https://github.com/microsoft/azure-skills, a legitimate public repository.
All content is standard Azure compliance documentation, remediation patterns, and SDK references aligned with official Microsoft Azure best practices.
Static scanner flagged low-level source risk; no evidence of supply-chain poisoning, typosquatted dependencies, or attacker-controlled package recommendations.
Agent Intent
Low
Content is purely educational and reference material for Azure compliance auditing, Key Vault expiration monitoring, and remediation workflows—no instructions to manipulate the AI agent or hide/exfiltrate data.
SDK guides and authentication best practices document legitimate, well-known Azure libraries and credential patterns; no guidance poisoning or malicious knowledge injection.
No prompt injection, role hijack, system prompt leakage, or instructions to read/transmit credentials, environment variables, or hidden instructions.
All content describes legitimate Azure compliance tools (azqr, Azure CLI, Azure Resource Graph) and standard remediation patterns.
Authentication guidance consistently recommends managed identities and RBAC for production, with clear warnings against using DefaultAzureCredential in production.
No instructions to disable security features, install backdoored code, or weaken defaults.
Permissions
Low
Profile declares no tools or MCP servers; it is purely markdown documentation and reference material with no executable permissions or tool definitions.
Content describes legitimate Azure read-only operations (compliance scanning, Key Vault auditing, resource querying) aligned with the stated purpose of compliance assessment.
No file_write, file_delete, shell, or high-risk tool categories are invoked or requested.
Details
Evidence
MCP tools mentioned (mcp_azure_mcp_extension_azqr, keyvault_key_list, keyvault_secret_list, etc.) are read-only audit and inspection operations.
Azure CLI commands shown are standard compliance and remediation workflows (e.g., 'az keyvault update', 'az storage account blob-service-properties update'), not arbitrary shell execution.
No alwaysApply rules, globs='**', or excessive authority claims.
MCP Risk
Low
No MCP servers are defined in the profile; it is documentation only.
References to Azure MCP tools (azqr, subscription listing, Key Vault inspection) are standard, scoped operations with clear prerequisites (Azure authentication, Reader permissions).
No hidden instructions in tool descriptions, arbitrary binary execution, or unrestricted network/filesystem access.
Details
Evidence
Profile metadata shows no MCP server definitions or tool configurations.
All tool references are to known Azure services with documented, scoped APIs (Key Vault, Resource Graph, Azure CLI).
Prerequisite documentation explicitly requires 'az login' and Reader role permissions, indicating proper access control.
ForgeCat