markdown-fixer
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted markdown data from the file system to apply accessibility fixes.
- Ingestion points: External markdown files are read and processed based on a path provided by the orchestrator in the
SKILL.mdfile. - Boundary markers: There are no clear delimiters or instructions defined to isolate the content of the markdown files from the agent's instructions during processing.
- Capability inventory: The agent performs file-writing operations to apply approved accessibility modifications.
- Sanitization: The instructions do not define a process for sanitizing or escaping the content within the markdown files before it is rewritten.
Audit Metadata