design-lark-chart
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads and executes @larksuite/whiteboard-cli via npx. This tool is provided by a well-known service (Lark/Feishu) and is used for diagram validation and rendering.
- [COMMAND_EXECUTION]: The skill executes lark-cli and whiteboard-cli commands to query whiteboard data, extract styles, and update whiteboard content on the Feishu platform.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted natural language input and external Feishu documents to plan and render charts.
- Ingestion points: The Normalize stage takes natural language or document URLs as input (SKILL.md).
- Boundary markers: The Planner instructions do not explicitly mandate the use of delimiters to encapsulate untrusted user content.
- Capability inventory: The skill has the capability to write directly to Feishu whiteboards via CLI tools.
- Sanitization: No explicit sanitization or instruction-filtering logic is described before passing data to the LLM-based Planner.
Audit Metadata