lark-workflow-task-prioritizer
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources, specifically Lark task summaries and calendar event descriptions fetched via the
lark-clitool. These inputs could potentially contain malicious instructions intended to manipulate the agent's output or behavior. - Ingestion points: Data entering the context includes output from
lark-cli task +get-my-tasks,+get-related-tasks,+search, andlark-cli calendar +agenda. - Boundary markers: None identified in the prompt to delineate or warn the agent about potentially malicious instructions within the task or calendar content.
- Capability inventory: The skill uses subprocess calls to execute
lark-clifor data retrieval. - Sanitization: No specific sanitization or filtering of the content from the external sources is performed before the agent analyzes it.
- [EXTERNAL_DOWNLOADS]: The skill contains references to external documentation and shared skill files hosted on the
larksuiteGitHub organization repository. These are identified as originating from a well-known technology service provider.
Audit Metadata