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

mcp-todoist: why it is not safe to use

mcp-todoist passed review: we found expected Todoist token use, no hidden prompts or hardcoded secrets, and documented destructive actions with dry-run support.

What we found

Our review of `greirson/mcp-todoist` passed with a score of 88/100. In the inspected repository materials, we did not find hidden prompt-injection content, obfuscated payloads, hardcoded credentials, privilege-escalation behavior, or undisclosed remote code download. The documented data flow is limited to the user-selected Todoist account through the Todoist API, and the code appears to use the expected `TODOIST_API_TOKEN` environment variable rather than unrelated local credentials.

We also noted that the skill supports actions that can change or remove user data, including `delete`, `bulk_delete`, `complete`, `archive`, and backup download. Based on the reviewed documentation, these capabilities are part of the stated Todoist functionality, and we did not find evidence of hidden destructive behavior. The presence of a dry-run mode is a useful safety control for testing.

What to do instead

Use this skill with a Todoist token you control, scope access to the intended account, and test workflows in dry-run mode before running bulk or destructive actions. Review the command set carefully so deletions, archiving, and completion actions match your intent.

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

Source: https://github.com/greirson/mcp-todoist

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.