jinko-data-table
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes command-line interfaces provided by the jinko-sdk, specifically jinko.cli.create_data_table and jinko.cli.inspect_data_table, for data processing and validation tasks.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from CSV, SQLite, and pandas DataFrames, which presents an ingestion surface for indirect prompt injection. * Ingestion points: External data files processed in SKILL.md through SDK upload functions. * Boundary markers: Explicit instructions to use assets/data-table.json for schema enforcement. * Capability inventory: SDK upload capabilities and metadata validation. * Sanitization: Automated dry-run row validation and interval logic checks are employed to mitigate risks from malformed input.
- [SAFE]: The skill uses official vendor-provided SDK methods and CLI tools to interact with the Jinko platform. It includes structured validation steps to ensure data integrity during upload.
Audit Metadata