add-onedrive

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • Indirect Prompt Injection Surface: The skill enables the agent to read and process content from external OneDrive files, which introduces a surface where data could potentially influence agent behavior.
  • Ingestion points: File content is accessed through the OneDriveForBusinessService methods in SKILL.md.
  • Boundary markers: The instructions do not define specific delimiters to isolate external file content from the agent's core instructions.
  • Capability inventory: The agent is configured with tools such as Bash, LSP, and file system modification (Write/Edit), which could be targets for instructions embedded in external data.
  • Sanitization: No explicit sanitization or validation logic is provided for the retrieved file content.
  • Command Execution: The skill performs routine environment configuration and project build operations using CLI tools.
  • Evidence: The use of pa app add data-source and npm run build in SKILL.md.
  • Context: These commands are standard for a Power Apps development workflow and are used to add data sources and verify code integrity.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 10:15 AM
Security Audit — agent-trust-hub — add-onedrive