AgentPod is building a private, secure device for your AI agent.The AgentPod device is coming.Coming soonBe first
We checked this and rejected itsecurity

Todoist AI: why it is not safe to use

Todoist AI passed our review with disclosed network use, no hardcoded secrets, and no signs of hidden instructions or privilege abuse.

What we found

Our review of Todoist AI did not identify a security or privacy issue that would block use. The repository passed our checks for hidden instructions, hardcoded secrets, obfuscation, undisclosed remote code fetches, credential harvesting, and privilege escalation. Network destinations we observed were disclosed and matched the stated purpose: Todoist through the Todoist SDK and the named hosted MCP endpoint at `ai.todoist.net` for OAuth and MCP access.

The skill can create, modify, and delete Todoist data, which is expected for this category. In the reviewed code and documentation, we did not find evidence of covert data transfer outside Todoist. The README also advises users to verify which Todoist account is connected before write operations.

What to do instead

Use Todoist AI as a normal Todoist-connected skill, with the usual care for write access. Confirm the connected Todoist account before allowing changes or deletions, and provide only the Todoist token or OAuth access needed for this integration. If you need stricter controls, prefer read-only workflows or limit use to accounts and projects appropriate for automated task management.

Want the same outcome, safely? Use our checked skill instead.

Source: https://github.com/doist/todoist-ai

We report what our security review found at the time we checked, with the goal of keeping people safe. Projects change; if a maintainer has since fixed this, we are glad to recheck it. Email hello@agentpod.com.

Copied to clipboard. Paste it into your AI (ChatGPT, Claude, or your agent) to add the skill.