feishu-card
Pass
Audited by Gen Agent Trust Hub on Mar 7, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill consists entirely of Markdown instructions and JSON templates. There are no executable scripts, shell commands, or external package dependencies provided within the skill files.
- [INDIRECT_PROMPT_INJECTION]: The JSON templates in the 'references/' directory use placeholders like '{{description}}' and '{{actions_list}}' for dynamic content. While this creates a surface for indirect prompt injection if untrusted data is used to fill these fields, the risk is localized to content rendering within the Feishu client and is considered low. ● Ingestion points: Template placeholders in 'card-alert.json', 'card-morning-briefing.json', and 'card-skill-test.json'. ● Boundary markers: None implemented in the provided templates. ● Capability inventory: The skill defines the structure for API calls to Feishu's message endpoints but does not perform the calls itself. ● Sanitization: Not specified in the documentation; implementers should sanitize data before interpolation.
Audit Metadata