ae-dataops
Pass
Audited by Gen Agent Trust Hub on May 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONCREDENTIALS_UNSAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill provides the ability to create and execute shell scripts via the ae-cli dataops_flow command with the SHELL task type. This allows the agent to run arbitrary commands on the target platform. It also provides the ability to execute arbitrary SQL, including DDL commands, on external databases via dataops_integration +execute_sql.\n- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection (Category 8) as it handles user-provided SQL, shell scripts, and configurations.\n
- Ingestion points: Untrusted data enters the agent context through SQL query strings in dataops_ide +execute_sql and dataops_integration +execute_sql, shell script content in dataops_flow +save_task_definition, and connection configurations in dataops_integration +add_datasource.\n
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands are provided for the agent to use when interpolating external data.\n
- Capability inventory: The skill possesses extensive capabilities, including executing SQL and DDL on data warehouses and external databases, running shell scripts on orchestration nodes, and managing data sync solutions.\n
- Sanitization: There is no evidence of input validation, escaping, or filtering of external content before it is passed to CLI commands.\n- [CREDENTIALS_UNSAFE]: Instructions for adding datasources involve passing sensitive database credentials, such as usernames and passwords, directly as command-line arguments. This practice can expose secrets to other users on the system via process monitoring tools or command history logs.
Audit Metadata