create-connections
Pass
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: Uses the
revosCLI to manage connections and retrieve source details. This is the intended functionality for the RevOS platform integration. - [DATA_EXFILTRATION]: The skill includes logic to detect sensitive fields (e.g., email, SSN) in source data and recommends using hashing or filtering to prevent plaintext exposure in the data warehouse.
- [PROMPT_INJECTION]: Identified an indirect prompt injection surface where the agent processes metadata from external sources.
- Ingestion points: Data enters via the output of
revos sources list-streams. - Boundary markers: No explicit delimiters are used to isolate metadata.
- Capability inventory: The skill has file-writing and CLI execution capabilities.
- Sanitization: Names from external sources are not sanitized before being incorporated into prompts or configuration files.
Audit Metadata