The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes content from Todoist tasks, projects, and comments, creating a potential attack surface where external data could contain instructions meant to influence the AI agent.
Ingestion points: External data enters the agent context through commands such as td task list, td task view, and td comment list described in SKILL.md.
Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded prompts within the retrieved task data.
Capability inventory: The agent is provided with CRUD capabilities via the td CLI and network access through direct API endpoints as seen in references/completed-tasks.md.
Sanitization: The skill does not provide methods for sanitizing or validating content retrieved from Todoist before processing.
[EXTERNAL_DOWNLOADS]: The skill includes instructions to install the @doist/todoist-cli package from the official npm registry, which is the recognized tool for the skill's intended purpose.

todoist-api