blog-writer
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied summaries and titles to generate article markdown and Astro page components. It lacks explicit boundary markers or sanitization instructions for this input, which creates a potential surface for indirect prompt injection if the user-supplied data contains malicious instructions. However, the risk is low due to the prescriptive nature of the writing templates and its focus on content generation.\n
- Ingestion points: User-provided article title, summary, and
references/topic-ideas.mdinSKILL.md.\n - Boundary markers: Absent in the workflow steps.\n
- Capability inventory: Writing to
docs/andsrc/pages/directories, updating_catalog.json, and executingnpm run buildthrough the shell.\n - Sanitization: No explicit sanitization or validation of the user-provided text before it is used to generate site files.\n- [COMMAND_EXECUTION]: The skill requires the execution of
npm run buildin Step 8 to verify that the generated blog articles and pages do not break the website's build process. This is a standard and expected operation for an Astro development environment and aligns with the primary purpose of the skill.
Audit Metadata