de-ai-ify
Pass
Audited by Gen Agent Trust Hub on Mar 8, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface due to its data-handling design. * Ingestion points: The skill processes external file content (.md, .txt, .docx) provided via the file_path argument. * Boundary markers: No delimiters are specified in SKILL.md to separate the processed text from the agent's instructions. * Capability inventory: The skill is tasked with reading, regex matching, and writing files. * Sanitization: No sanitization or validation of the input content is described in the documentation.
- [NO_CODE]: No executable code is provided in the skill files. * Missing scripts: The skill relies on the agent's execution of instructions provided in Markdown format; no Python, JavaScript, or other scripts were found.
Audit Metadata