markdown-plan

Pass

Audited by Gen Agent Trust Hub on Jul 22, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill ingests untrusted user data and uses it to generate both filenames and file content without safety boundaries or sanitization.
  • Ingestion points: The user's request is used to determine the name and body of the markdown plan file.
  • Boundary markers: Absent; there are no instructions to delimit user-provided data or to ignore embedded instructions within the user input.
  • Capability inventory: The skill utilizes file-writing capabilities within the user's home directory (~/.cursor/plans/).
  • Sanitization: No validation is performed on the filename to prevent directory traversal (e.g., using '../../' in the name) and no escaping or filtering is applied to the markdown body.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 22, 2026, 12:38 AM
Security Audit — agent-trust-hub — markdown-plan