lark-workflow-bitable-sync
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from external Lark/Feishu Bitables, which could potentially contain malicious instructions intended to influence the agent's behavior during the mapping or sync process.
- Ingestion points: Data is retrieved from external tables using
lark-cli base +record-listas described in Step 3 and Step 4 ofSKILL.md. - Boundary markers: The instructions lack explicit boundary markers or "ignore embedded instructions" directives to prevent the AI from following commands hidden within the table cell content.
- Capability inventory: The skill has the capability to write and modify data in the target system using
lark-cli base +record-upsertin Step 6. - Sanitization: There is no evidence of sanitization, validation, or escaping of the external table content before it is processed by the AI model or used to construct JSON payloads for writing.
Audit Metadata