content-research-writer
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates as a purely instructional assistant for content creation. It does not contain any code, scripts, or remote dependencies.
- [DATA_EXPOSURE]: The skill instructions guide the user to create a project directory in
~/writing/. This is a benign organizational practice and does not target sensitive system paths, credentials, or private configuration files. - [COMMAND_EXECUTION]: The provided examples use standard, non-privileged shell commands (
mkdir,touch,cd) for file management. These are intended for user-driven setup and do not represent autonomous or malicious execution. - [INDIRECT_PROMPT_INJECTION]: The skill processes user drafts and external research data. While this is an ingestion surface for potential indirect instructions from web sources, the skill lacks high-impact capabilities (such as arbitrary code execution or network exfiltration of sensitive files) that would allow for exploitation. The behavior is consistent with safe writing assistance.
Audit Metadata