The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it is designed to ingest and process data from external Snowflake databases which could contain malicious instructions. \n
Ingestion points: Results from SQL queries (e.g., SELECT * FROM MY_TABLE in SKILL.md) are returned to the agent context.\n
Capability inventory: The skill utilizes shell commands (curl, jq) for network operations and data manipulation.\n
Boundary markers: The provided examples do not implement delimiters or instructions to ignore potential commands embedded within the retrieved data.\n
Sanitization: There is no evidence of sanitization or validation applied to the queried data before it is processed by the agent.\n- [COMMAND_EXECUTION]: The skill uses curl to perform API requests and jq to process JSON output. These are standard utilities used for their intended purpose in a developer-oriented skill.\n- [DATA_EXFILTRATION]: Performs network requests to snowflakecomputing.com. As this is a well-known cloud service, the network operations are documented and necessary for the skill's primary function.\n- [SAFE]: The skill references zero doctor, which appears to be a vendor-provided diagnostic utility for troubleshooting connector status.

snowflake