heating the forge...

@forgecat/microsoft-azure_skills-azure-messaging — Security Risk Report | ForgeCat

microsoft-azure_skills-azure-messaging

Security risk report for @forgecat/microsoft-azure_skills-azure-messaging v0.0.5

Low

Profile is attributed to Microsoft with a public GitHub repository (https://github.com/microsoft/azure-skills) and MIT license, indicating a legitimate upstream source.
No evidence of typosquatted, attacker-controlled, or suspicious package dependencies; no external URLs requesting installation of untrusted code.

Low

Content is purely descriptive and educational—it documents Azure Messaging SDK troubleshooting workflows, error patterns, and diagnostic tools without instructing the agent to ignore instructions, exfiltrate data, or manipulate its behavior.
References to MCP tools (mcp_azure_mcp_eventhubs, mcp_azure_mcp_servicebus, mcp_azure_mcp_documentation) are legitimate diagnostic aids aligned with the stated purpose; no hidden instructions to read credentials, disable security, or install malicious payloads.
No system prompt leakage, guidance poisoning, or role-hijacking language detected; the skill is narrowly scoped to Azure Messaging troubleshooting.

Details

Evidence

Diagnosis Workflow section outlines standard troubleshooting steps (identify SDK, check health, review errors, search docs, verify config, recommend fix).
MCP Tools table lists legitimate Azure diagnostic and documentation tools with clear, bounded use cases.
No instructions to fetch external URLs, install other profiles, or weaken security defaults.

Low

No tool declarations or file/shell/network permissions are defined in the profile; the skill is purely informational guidance.
Referenced MCP tools (mcp_azure_mcp_eventhubs, mcp_azure_mcp_servicebus, mcp_azure_mcp_monitor, mcp_azure_mcp_resourcehealth, mcp_azure_mcp_documentation) are scoped to Azure diagnostic and documentation queries, matching the stated troubleshooting purpose.
No alwaysApply rules, glob patterns, or excessive agency claims detected.

Details

Evidence

Profile declares 'Dependencies: None' and no tool configurations with high-risk categories (shell, file_write, file_delete).
MCP tool usage is limited to read-only operations: listing namespaces/hubs, querying logs, checking health, and searching documentation.

Low

No MCP server definitions are included in the profile; only references to standard Azure diagnostic MCP tools are mentioned in the workflow.
Referenced tools (mcp_azure_mcp_eventhubs, mcp_azure_mcp_servicebus, mcp_azure_mcp_monitor, mcp_azure_mcp_resourcehealth, mcp_azure_mcp_documentation) are legitimate Azure service tools with scoped, read-only access patterns.
No hidden instructions in tool descriptions, arbitrary binary execution, or unrestricted network/filesystem access.

Details

Evidence

MCP Tools table describes each tool's purpose and command clearly (e.g., 'Namespace/hub ops', 'logs_query', 'Doc search').
No environment credentials, secrets, or external host access requirements declared.