Security audit
Security checks for vulnerabilities and agentic risk
This Todoist skill is a straightforward CLI reference for managing a Todoist account, with expected account-changing commands but no hidden execution or persistence.
Install only if you are comfortable giving the CLI access to your Todoist API token. Before deleting tasks, projects, sections, labels, or comments, have the agent show the exact target and ask for explicit confirmation.
Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.
# Get single task
todoist task TASK_ID
# Add task
todoist add "Buy groceries"
todoist add "Call mom" --due tomorrow
todoist add "Meeting prep" --due "today 3pm" --priority 4
The skill documents task deletion and state-changing commands such as complete, reopen, and delete without any guidance to confirm user intent or warn that deletion may be irreversible. In an agent context, this increases the chance of accidental destructive actions against a real user account, especially when a model may translate ambiguous requests into direct CLI execution.
The skill exposes deletion commands for projects, sections, labels, and comments with no user-facing warnings about scope, permanence, or confirmation requirements. Because these commands affect organizational data and may remove large sets of associated task structure, an agent using this skill could cause unintended data loss from a single mistaken or maliciously prompted action.
No suspicious patterns detected.