naver-engage
Pass
Audited by Gen Agent Trust Hub on Nov 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates an automated loop that consumes untrusted data from external blog posts to generate and publish comments, creating an attack surface for indirect prompt injection.\n
- Ingestion points: The
neighbor_post_commenttemplate instructions describe reading new posts from a neighbor's feed (SKILL.md).\n - Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from obeying instructions embedded within the processed blog posts.\n
- Capability inventory: The skill possesses the capability to perform automated writes (comments) and other social actions like following and liking on the Naver platform via the Uplika bridge.\n
- Sanitization: The skill lacks mechanisms to sanitize or filter the content of the blog posts before they are processed by the AI for comment generation.
Audit Metadata