risk-adjusted-return-optimizer
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill relies on gathering user-specific data (portfolio size, risk preference, investment term, and constraints) to generate a structured financial recommendation. This creates an ingestion surface for potentially malicious instructions hidden within user inputs, which could be designed to influence the agent's logic.\n
- Ingestion points: User-provided inputs collected in Step 1 of
SKILL.md(e.g., amount, risk, horizon, and account-specific constraints).\n - Boundary markers: The instructions do not define explicit delimiters or "ignore embedded instructions" warnings to separate user data from the core logic.\n
- Capability inventory: The skill refers to
findata-toolkit-cnfor real-time market data retrieval; no dangerous system-level execution or unauthorized network exfiltration capabilities were identified.\n - Sanitization: There are no mentioned mechanisms for validating, filtering, or escaping the content provided by the user before it is interpolated into the final output.
Audit Metadata