skills/larksuite/cli/lark-task/Gen Agent Trust Hub

lark-task

Pass

Audited by Gen Agent Trust Hub on Oct 29, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes and displays task content (summaries, descriptions, and comments) retrieved from an external service (Lark). This data could contain malicious instructions designed to influence the agent's behavior when it parses and renders task lists or details.
  • Ingestion points: Task summaries, descriptions, and comments are read through commands like +search, +get-my-tasks, and tasks.get (SKILL.md, references/lark-task-search.md).
  • Boundary markers: The instructions provide guidelines for rendering (formatting timestamps and resolving user IDs) but do not explicitly instruct the agent to ignore instructions embedded within the task text itself.
  • Capability inventory: The agent has the ability to execute shell commands via lark-cli and upload local files via +upload-attachment (references/lark-task-upload-attachment.md).
  • Sanitization: No explicit sanitization or escaping of task content is mentioned before it is processed or presented to the user.
  • [COMMAND_EXECUTION]: The skill's primary function is to execute the lark-cli binary to interact with the Lark Task API. The agent is instructed to dynamically discover subcommands and parameters using lark-cli task --help and lark-cli schema, which is a standard and expected behavior for CLI-based skills.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 29, 2026, 08:53 PM