markdown-utilities
Pass
Audited by Gen Agent Trust Hub on Jun 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
md-utilscommand-line tool to perform operations such as reading frontmatter, generating tables of contents, and extracting sections from Markdown files.\n- [PROMPT_INJECTION]: The skill exposes an indirect prompt injection surface because it enables the agent to read and process content from external, potentially untrusted, Markdown documents.\n - Ingestion points: Data is ingested into the agent context via commands like
md-utils explore,md-utils extract, andmd-utils body, as described inSKILL.md.\n - Boundary markers: The instructions lack specific markers or warnings to prevent the agent from following instructions that might be embedded within the files being read.\n
- Capability inventory: The agent has the capability to write to the file system (modifying YAML frontmatter) using the
md-utils fm setcommand.\n - Sanitization: No input validation or sanitization is specified for the Markdown content before it is read or manipulated by the agent.
Audit Metadata