tracking-crypto-derivatives
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides legitimate functionality for monitoring cryptocurrency markets. The implementation is highly structured, modular, and adheres to standard security practices for data normalization using specific types like Decimal and float.\n- [INDIRECT_PROMPT_INJECTION]: The skill's architecture relies on fetching and processing data from external exchange APIs (e.g., Binance, Bybit, OKX). This creates a surface for indirect prompt injection if an attacker were to manipulate the data returned by these services to influence the agent's behavior.\n
- Ingestion points: External exchange data is fetched in
scripts/exchange_client.pyand analyzed in specialized scripts likefunding_tracker.pyandoi_analyzer.py.\n - Boundary markers: Absent. The formatted reports (tables and dashboards) do not utilize explicit delimiters to separate external content from system instructions.\n
- Capability inventory: The skill has access to
Bash(scoped to project scripts) and file system operations (Read,Write,Edit,Grep,Glob).\n - Sanitization: Values are normalized into dataclasses and cast to specific numeric types (
Decimal,float), providing a basic layer of input validation.\n- [SAFE]: No malicious obfuscation, hardcoded secrets, or persistence mechanisms were detected. All external references are to established financial platforms or the author's verified professional domains.
Audit Metadata