ave-data-wss
Pass
Audited by Gen Agent Trust Hub on Mar 18, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests live blockchain data (swaps, prices, and candle updates) which is external and potentially attacker-controlled content. \n
- Ingestion points: Data is fetched from the vendor's WebSocket API at wss://wss.ave-api.xyz via scripts/ave_data_wss.py. \n
- Boundary markers: There are no explicit instructions or delimiters defined to separate untrusted data from agent instructions. \n
- Capability inventory: The skill executes local Python scripts and manages Docker containers, providing a broad capability set. \n
- Sanitization: No sanitization or validation of the incoming blockchain data is mentioned in the operational guidelines. \n- [COMMAND_EXECUTION]: The skill requires running local Python scripts and Docker commands to manage connection daemons and REPL environments, which are necessary for its real-time data streaming functionality. \n- [EXTERNAL_DOWNLOADS]: The installation process involves downloading dependencies via pip from standard package registries.
Audit Metadata