md-upgrade
Pass
Audited by Gen Agent Trust Hub on Jul 25, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted data from external markdown files.
- Ingestion points: The skill reads content from a file path specified by the user or identified in the environment (SKILL.md).
- Boundary markers: There are no explicit delimiters or instructions to ignore instructions embedded within the target markdown file.
- Capability inventory: The skill utilizes file read and Edit/Write capabilities to modify the target document (SKILL.md).
- Sanitization: The skill references Mermaid-specific safety rules (quoted labels, node limits) but lacks general sanitization for the markdown content being processed.
- [COMMAND_EXECUTION]: The skill performs file read and write operations on paths provided by the user or identified in the environment. While central to its function, this capability requires the agent to validate file paths to prevent access to sensitive files outside the intended project scope.
Audit Metadata