The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes Python scripts (todo_batch_create.py, todo_daily_summary.py, todo_overdue_check.py) that execute the dws CLI via subprocess.run(). This is the intended mechanism for the skill to interact with the DingTalk environment. The scripts use list-based command construction, which prevents shell injection vulnerabilities.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data that could contain malicious instructions designed to influence the agent's behavior.
Ingestion points: The agent ingests data from DingTalk via dws todo task list and dws aisearch person, and from local JSON files provided by the user to todo_batch_create.py.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to distinguish between task content and its own operating instructions.
Capability inventory: The skill possesses write capabilities through dws todo task create, update, and delete commands, as well as local shell execution capabilities via the included scripts.
Sanitization: While todo_batch_create.py validates fields like priorities and IDs, it does not explicitly sanitize or escape string content such as task titles before they are processed by the agent or passed to tools.

dingtalk-todo