setup-sakana-skills
Pass
Audited by Gen Agent Trust Hub on Nov 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes project-level configuration files, which presents a potential surface for indirect prompt injection if those files contain adversarial instructions.
- Ingestion points: The skill reads instructions and configuration from
AGENTS.md,CLAUDE.md, and various documents within thedocs/agents/directory (found inSKILL.md). - Boundary markers: No explicit delimiters or boundary markers are defined to isolate content ingested from these external configuration files to prevent the agent from obeying embedded instructions.
- Capability inventory: The skill has the capability to write to project configuration files to establish pointers to guidance documents, as described in the "Apply and verify" sections.
- Sanitization: No specific sanitization or validation of the ingested configuration content is described beyond the instruction to preserve existing guidance and resolve missing choices.
Audit Metadata