skills/microsoft/hve-core/mural/Gen Agent Trust Hub

mural

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFE
Full Analysis
  • Industry-Standard Credential Management: The skill implements secure storage for OAuth tokens and application credentials using OS-level keychains (via the keyring package) or local files with restricted 0600 permissions. The security model includes a full STRIDE threat analysis, demonstrating a high level of security maturity.
  • Comprehensive Input Validation: The implementation includes rigorous validation for Mural identifiers, SSRF protection via hostname allowlisting for asset uploads to Azure Blob Storage, and strict scheme allowlists for hyperlinks to prevent cross-tenant attack vectors.
  • Privacy-Focused Logging: A dedicated redaction mechanism is utilized to scrub sensitive information, including OAuth tokens, client secrets, and PKCE parameters, from all log output and error messages.
  • Content Security Awareness: The documentation explicitly identifies user-authored content from Mural boards as untrusted input and provides guidance for downstream AI agents to handle this data safely, mitigating risks associated with indirect prompt injection surfaces.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 02:35 PM
Security Audit — agent-trust-hub — mural