todoist-2

Warn

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of a third-party binary 'todoist' from an unverified GitHub repository (buddyh/todoist-cli) via Homebrew or Go, which does not originate from a verified or well-known organization.
  • [COMMAND_EXECUTION]: The skill's primary functionality is based on executing shell commands through the 'todoist' CLI to interact with the Todoist API.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from Todoist (task names, descriptions, and comments) which constitutes an external, untrusted input source. 1. Ingestion points: Reading task lists ('todoist tasks'), viewing specific tasks ('todoist view'), and searching ('todoist search') ingest data from the Todoist API into the agent's context. 2. Boundary markers: There are no explicit delimiters or instructions to distinguish between task data and agent instructions in the provided documentation. 3. Capability inventory: The skill can execute shell commands via the 'todoist' CLI and perform write operations to the remote service. 4. Sanitization: There is no evidence of sanitization, filtering, or validation of the task content retrieved from the API before it is processed by the agent.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 2, 2026, 05:36 AM
Security Audit — agent-trust-hub — todoist-2