readme-beautifier
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from user-provided README files, which is an inherent part of its functionality.
- Ingestion points: Reads and reformats markdown documents (e.g., README.md, CONTRIBUTING.md) provided in the skill context.
- Boundary markers: There are no specific delimiters defined to isolate the user-provided text from the agent's internal instructions.
- Capability inventory: The skill definition contains no file writing, shell command execution, or network communication capabilities.
- Sanitization: No validation or filtering is applied to the input text, though the instructions explicitly forbid adding or removing content.
- [SAFE]: The skill follows best practices by restricting its scope to formatting and structure, explicitly forbidding changes to technical content, commands, or logic.
Audit Metadata