write-blog-post
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by ingesting and processing untrusted external data.
- Ingestion points: The skill instructions in
SKILL.mdexplicitly direct the agent to "turn notes, a transcript or a social post into a blog post" and to use "facts the user gives you or that are in their source." - Boundary markers: There are no instructions to use delimiters or specific boundary markers to separate untrusted source content from the agent's instructions, nor are there warnings to ignore embedded commands.
- Capability inventory: The skill possesses significant capabilities, including the ability to upload media from arbitrary URLs (
upload_media_from_url), create and modify web content (create_post,update_post), and publish content live to the internet (schedule_post). - Sanitization: The instructions do not define any sanitization, filtering, or validation steps for the data retrieved from external transcripts or social posts before it is processed or published.
Audit Metadata