writing-shape
Pass
Audited by Gen Agent Trust Hub on Jul 16, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection through the processing of untrusted input data.
- Ingestion points: The skill reads a user-provided markdown file (referred to as 'raw material' or the 'pile') from the filesystem in SKILL.md.
- Boundary markers: There are no defined delimiters or 'ignore' instructions intended to isolate the content of the pile from the agent's operational logic, which could allow malicious instructions in the text to hijack the task.
- Capability inventory: The skill possesses the capability to read local files and write or append content to files on the disk (SKILL.md defines a 'Writing rhythm' that involves disk operations).
- Sanitization: No sanitization or validation logic is specified for the data retrieved from the input file before it is interpreted by the agent during the article-shaping process.
Audit Metadata