style-journal-rewrite
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted external data that could contain malicious instructions. 1. Ingestion points: The skill processes user-supplied drafts (.docx, .md, .txt) and external reference materials for journal formatting and writing styles. 2. Boundary markers: The skill includes instructions to strictly follow user materials and avoid fabrication, but does not implement robust structural delimiters to separate untrusted content from system instructions. 3. Capability inventory: The skill can read and write local files using python-docx and execute bundled scripts (convert_ref_format.py, remove_heading_numbers.py) and can invoke other agent skills for citation management. 4. Sanitization: The skill does not perform content-based filtering or sanitization of the input documents before processing.
- [COMMAND_EXECUTION]: The skill utilizes localized Python scripts to perform document structure and citation formatting updates. Scripts scripts/convert_ref_format.py and scripts/remove_heading_numbers.py are executed to modify DOCX files as part of the journal adaptation workflow.
- [SAFE]: The skill's external references and metadata do not indicate malicious intent. References to github.com/aipoch/medical-research-skills are author-owned vendor resources. The inclusion of a security evaluation report and a polish changelog are standard development artifacts, though their internal safety claims are treated as data rather than authoritative conclusions.
Audit Metadata