create-x-content
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [SAFE]: The skill performs legitimate content generation tasks by reading local configuration files and writing text output to a designated directory without network access or unauthorized command execution.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from a user-provided brief and local configuration files, which is inherent to its functionality as a content generator.
- Ingestion points: Untrusted brief data is accepted via the
--briefargument, and external voice guide files are read from~/.goose-skills/into the agent's context (SKILL.md). - Boundary markers: The instructions do not define explicit delimiters or 'ignore' instructions when interpolating the brief into the generation prompt.
- Capability inventory: The skill is restricted to reading its own configuration files and writing markdown files to the local filesystem (SKILL.md).
- Sanitization: No explicit validation or filtering of the brief content is performed before processing.
- [NO_CODE]: The skill is entirely agent-executed and does not include scripts, binaries, or runtime code execution.
Audit Metadata