todoist
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill interacts with the Todoist service via a local CLI binary named
todoist. This is the intended design for this integration. - [EXTERNAL_DOWNLOADS]: The skill performs network requests to official Todoist domains (
todoist.comandai.todoist.net) for authentication and task management. These are recognized endpoints for a well-known service. - [INDIRECT_PROMPT_INJECTION]: The skill possesses an ingestion surface for external content from Todoist (tasks, projects, comments) as described in
SKILL.md. While no specific sanitization or boundary markers are documented, the risk is mitigated by themanifest.yamlconfiguration, which requires explicit user consent ('ask' default) for all write and delete capabilities, ensuring human-in-the-loop review.
Audit Metadata