data-table-manager
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing external data from file attachments.
- Ingestion points: Processes data from CSV, XLSX, and JSON files using the
parse-filetool. - Boundary markers: Contains explicit instructions to treat all parsed values as untrusted data and not as instructions, and to preserve source values even when they resemble commands.
- Capability inventory: The skill uses
data-tablesto create, modify, and delete data rows. - Sanitization: Includes instructions to store potential spreadsheet formulas as plain values and to never evaluate them.
- [NO_CODE]: The skill does not contain any executable scripts or code files, relying solely on markdown instructions and platform tools.
- [SAFE]: No malicious patterns or security risks were identified in the skill instructions or metadata.
Audit Metadata