book-illustration-workflow
Pass
Audited by Gen Agent Trust Hub on Apr 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to use the
lark-cliutility to manage documents on the Feishu platform. It specifically uses commands likedocs +create,docs +update, anddocs +media-insertto automate the synchronization process. - [DATA_EXFILTRATION]: The core functionality involves uploading local content, including text from Markdown files and image assets, to the external Feishu cloud service. While this is the intended use case for syncing work, it represents a data transfer from the local environment to a remote service.
- [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection (Category 8).
- Ingestion points: The skill reads and processes local Markdown files (chapters) to identify placeholders and perform cleanup.
- Boundary markers: None identified in the instructions to separate untrusted content from agent logic.
- Capability inventory: The skill has file-write capabilities (modifying local Markdown) and network operations (uploading to Feishu via
lark-cli). - Sanitization: There is no explicit sanitization or validation of the content being processed before it is used to drive the synchronization workflow.
Audit Metadata