Coinbase Automation
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE] (SAFE): No malicious patterns detected. The skill's behavior is consistent with its stated purpose of cryptocurrency wallet management.
- [DATA_EXFILTRATION] (SAFE): While the skill processes sensitive financial metadata (wallet balances and IDs), it does so within the context of the requested Rube MCP server (https://rube.app/mcp). No secondary or unauthorized exfiltration channels were identified.
- [PROMPT_INJECTION] (SAFE): The skill does not contain instructions that attempt to bypass safety filters or override system prompts. It follows a standard instructional format.
- [INDIRECT_PROMPT_INJECTION] (SAFE): The skill processes external data (wallet metadata) via the
COINBASE_LIST_WALLETStool. - Ingestion points: Data enters the agent via the output of the
COINBASE_LIST_WALLETStool. - Boundary markers: None explicitly defined in the prompts.
- Capability inventory: The skill is limited to read-only operations for listing and monitoring wallets.
- Sanitization: None detected. Given the read-only scope, the risk is negligible.
Audit Metadata