markdown-formatter
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOW
Full Analysis
- [Indirect Prompt Injection] (INFO): The skill processes untrusted markdown content provided in the $ARGUMENTS variable.
- Ingestion points: $ARGUMENTS in SKILL.md.
- Boundary markers: None present around input interpolation.
- Capability inventory: Only text formatting and display. No file-write, subprocess, or network capabilities are present.
- Sanitization: No sanitization of input markdown.
- Risk Assessment: While an attacker could embed instructions in markdown, the skill has no dangerous tools or side effects to exploit. The risk is limited to the agent potentially following formatting-related instructions within the data.
- [Overall Assessment] (SAFE): No malicious patterns, obfuscation, or suspicious dependencies were found. The skill behavior matches its stated purpose.
Audit Metadata