The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Todoist CLI (td) to perform task management operations, verify task states, and diagnose authentication issues. These commands are executed within the agent's environment to supplement the MCP tools.
Evidence: instructions in references/todoist-mcp.md for using td today, td upcoming, and td task reschedule.
[EXTERNAL_DOWNLOADS]: The skill references the installation of the @doist/todoist-cli package from the official npm registry. This is a trusted resource from a well-known service provider.
Evidence: npm install -g @doist/todoist-cli in references/todoist-mcp.md.
[PRIVILEGE_ESCALATION]: The documentation notes that the td CLI utility may require platform-approved escalation to read authentication tokens from the macOS Keychain when operating within a sandbox.
Evidence: Reference to "approved escalation" for Keychain access in references/todoist-mcp.md.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external sources that could contain instructions, such as Todoist task descriptions and Google Calendar events.
Ingestion points: User input ("messy thoughts"), Todoist task titles and descriptions, and Google Calendar event details (SKILL.md, references/todoist-mcp.md).
Boundary markers: None identified; instructions are processed directly to triage tasks.
Capability inventory: The skill can create, update, and reschedule tasks via Todoist MCP and CLI, and execute arbitrary commands via codex exec (references/todoist-mcp.md).
Sanitization: No specific sanitization or filtering of external task content is documented.

personal-task-system