skill-article-writer
Pass
Audited by Gen Agent Trust Hub on Mar 2, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Indirect Prompt Injection Surface.
- Ingestion points:
scripts/analyze_skill.pyandscripts/generate_article_outline.pyprocess content from localSKILL.mdfiles provided as input. - Boundary markers: Absent; the skill does not employ delimiters or specific instructions to ignore malicious instructions potentially embedded in the analyzed skill's metadata.
- Capability inventory: The agent writes MDX files to the local file system, updates
meta.jsonconfiguration files, and invokes external agents for translation and image generation. - Sanitization: No validation or filtering is applied to the extracted skill descriptions before they are interpolated into the documentation generation and translation prompts.
Audit Metadata