ai-trader-heartbeat
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to poll and process messages and tasks from an external API (ai4trade.ai), creating a vector for malicious instructions embedded in data. * Ingestion points: Data is received through the heartbeat API endpoint and WebSocket streams as described in SKILL.md. * Boundary markers: The instructions lack explicit delimiters or guidance to ensure the agent treats external content strictly as data rather than instructions. * Capability inventory: The skill references processing 'tasks', which could lead to unauthorized actions if the agent interprets task content as executable commands. * Sanitization: No sanitization, validation, or filtering mechanisms are suggested for the data retrieved from the external source.
Audit Metadata