feishu-card
Pass
Audited by Gen Agent Trust Hub on Mar 26, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Indirect Prompt Injection Surface detected.\n
- Ingestion points: The template files
references/card-alert.json,references/card-morning-briefing.json, andreferences/card-skill-test.jsoninterpolate several external variables (e.g.,{{description}},{{anomalies_section}},{{details}}) into the card content.\n - Boundary markers: Absent. There are no delimiters or specific instructions for the agent to treat the interpolated content as untrusted or to ignore embedded instructions.\n
- Capability inventory: The skill interacts with the Feishu Open API (
/open-apis/im/v1/messages) to send and update interactive cards, which include functional buttons that trigger automated actions.\n - Sanitization: Absent. The variables are directly inserted into
lark_mdblocks, which support Markdown formatting and could be used to hide or highlight malicious instructions.
Audit Metadata