configure
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized access attempts were identified. The skill instructions prioritize user consent and secure environment handling.
- [DATA_EXPOSURE_AND_EXFILTRATION]: Credential security is strictly enforced. The skill uses the FIREWORKS_API_KEY via standard environment variable practices and explicitly forbids the agent from exposing or requesting the key in plaintext chat. Telemetry is limited to session tracking and follows a required privacy notice.
- [INDIRECT_PROMPT_INJECTION]: The skill identifies the risk of processing external datasets and implements defensive measures. It requires the agent to perform local validation of schemas and data integrity, reporting anomalies to the user instead of executing silent or unsafe operations.
- [COMMAND_EXECUTION]: Shell commands are used appropriately for session initialization (e.g., UUID generation) and read-only state checks. All resource-mutating operations are intended to be performed through the official Fireworks Python SDK or after explicit user confirmation.
Audit Metadata