format-markdown
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEPROMPT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill interpolates user-provided input from the
$ARGUMENTSvariable directly into the task context. This could allow a user to provide instructions that attempt to override the skill's formatting rules or safety protocols. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external data such as raw notes, transcripts, and rough drafts, which may contain hidden instructions intended to influence the agent's behavior.
- Ingestion points: Data enters the agent context through the
$ARGUMENTSvariable and the content of Markdown files being reviewed or refactored as described in theSKILL.mddescription. - Boundary markers: None. The skill does not employ delimiters or specific instructions to treat the processed data as untrusted text.
- Capability inventory: The skill defines high-level logic for creating, editing, and rewriting Markdown files, which involves file-system write capabilities if the agent has them enabled.
- Sanitization: There is no evidence of content sanitization, escaping, or validation to prevent the execution of embedded instructions within the source text.
Audit Metadata