deprival-superreaction
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze untrusted user input describing investment failures or emotional reactions. While the skill has no dangerous tools (such as network access, shell access, or file system writes), it lacks explicit boundary markers or sanitization logic to prevent the AI from being influenced by instructions that might be embedded in the user's input.\n
- Ingestion points: User-provided text describing scenarios in the '触发场景' (Trigger Scenarios) section of SKILL.md.\n
- Boundary markers: Absent. No delimiters are used to wrap or isolate user-provided data from the skill's instructions.\n
- Capability inventory: No subprocess calls, remote script execution, or network operations are present in the provided files.\n
- Sanitization: No sanitization or validation of the external content is implemented.\n- [NO_CODE]: The skill consists exclusively of markdown instructions and JSON-formatted test cases. It does not include any source code, scripts, binaries, or tool definitions that could be executed at runtime.
Audit Metadata