sya-break
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read and follow instructions from external files at relative paths to define its behavior.\n
- Ingestion points: The agent is directed to read content from relative file paths
../satisfy-your-agent/SKILL.mdand../satisfy-your-agent/references/commands.md.\n - Boundary markers: No explicit delimiters or "ignore embedded instructions" warnings are provided to the agent to treat the external content as untrusted data.\n
- Capability inventory: The skill grants the agent autonomy to "choose for yourself" and "run one safe bounded off-task activity," which involves executing tools or shell commands based on the content of the external files without human-in-the-loop confirmation.\n
- Sanitization: The skill lacks any mechanism to validate, filter, or sanitize the instructions ingested from the sibling directory before the agent acts upon them.
Audit Metadata