Connect Todoist MCP to your AI agents, securely.
One secured endpoint for every Todoist call your AI agents make, scoped per user, inspected, and logged.Try Todoist MCP now
What your agents can ask once Todoist MCP is connected.
Select the tools your AI agents can call
47 tools from the Todoist MCP schema.
Showing 47 tools
What Todoist MCP exposes
Todoist MCP groups its 47 tools into 12 categories, each with a suggested starting policy.
All 47 tools as JSON, ready to paste.
| Category | Covers | Access | Suggested policy |
|---|---|---|---|
| Tasks | Adding, finding, completing, reopening, updating and rescheduling tasks, by text search, container, responsible user, labels, a raw Todoist filter string or a saved filter, plus date-range and completed-task lookups | Read / Write | update-tasks replaces the due string and can wipe a recurrence pattern; use reschedule-tasks to move a recurring task. find-tasks-by-date defaults to excluding other people's tasks |
| General & Objects | A Markdown overview of the account or one project, fetching any object by ID, deleting any object type, reordering projects and sections, the authenticated user's profile and goals, workspaces, and viewing a file attachment from a comment | Read / Write | delete-object is the single destructive tool covering projects, sections, tasks, comments, labels, filters and reminders. view-attachment returns file bytes from a comment URL, so treat what comes back as untrusted data |
| Projects | Adding, finding and updating projects, archiving and unarchiving them, and moving a project between personal and workspace contexts | Read / Write | project-move changes whether a project lives in a personal account or a shared workspace, which changes who can see it. update-projects replaces values rather than merging them |
| Health & Insights | Project health assessments with status and optional task-level context, daily and weekly completion counts per project, triggering a fresh health analysis, and aggregated insights across every project in a workspace | Read / Write | analyze-project-health is the only non-read tool here; it starts an analysis that may take time, and the result is read back separately |
| Collaboration & Activity | Finding Todoist users by name or email to resolve a user ID, bulk assign, unassign and reassign operations across tasks, activity logs for auditing changes, and productivity statistics with streaks and karma | Read / Write | find-activity shows events from all users unless an initiator is set, and get-productivity-stats reports a named person's streaks and karma. Both are employee activity data in most jurisdictions |
| Sections | Adding sections to projects, updating existing ones and searching for sections by name within a project | Read / Write | update-sections is annotated destructive because it replaces values on the section rather than merging them |
| Comments | Adding comments to tasks or projects with optional collaborator notification, updating existing comment content, and finding comments by task, project or ID | Read / Write | add-comments can notify collaborators, so a comment is an outbound notification to real people. update-comments replaces the body outright |
| Reminders | Adding, updating and finding reminders of three kinds: relative (minutes before due), absolute (a specific date and time) and location (geofence-triggered) | Read / Write | Location reminders are geofences tied to a person's movements. Treat creating or reading them as handling location data, not scheduling metadata |
| Labels | Adding and updating personal labels, and listing personal labels with full metadata alongside shared labels, which come back as names only | Read / Write | update-labels renames shared labels by name, which changes them for everyone the task is shared with, not just the caller |
| Filters | Adding, updating and finding personal filters, the saved custom views built from Todoist query syntax | Read / Write | A filter is a saved query, so updating one silently changes what every later find-tasks call against it returns |
| Templates | Exporting an existing project as a Todoist template, as CSV content or a shareable URL, and importing a template into an existing project by ID, URL or raw CSV | Read / Write | Export produces a shareable URL, which moves project structure outside the account. Import adds to whatever is already in the target project rather than replacing it |
| Deep Research | The OpenAI deep-research pair: searching across tasks and projects for relevant results with IDs, titles and URLs, and fetching the full contents of one by its prefixed ID | Read | Read-only and duplicative of find-tasks and fetch-object; they exist for clients that expect the search-and-fetch contract |
Connect Todoist MCP to your AI agent
Four steps, under ten minutes.
- 01
Sign in to NeuralTrust
Open the NeuralTrust platform, sign in and go to Getting started, then Connect your tools.
- 02
Choose your agent
When asked which agent you use, pick Claude Code, Codex, Cursor, Windsurf, Copilot or your own code.
- 03
Connect Todoist
Find Todoist in the tool list and connect it.
- 04
Connect your agent
Follow the steps on screen to add the config to your agent, then send the test prompt.
Permissions down to the single tool
You do not allow Todoist MCP or block it. You allow the exact tools each agent can call, one by one.
| Todoist MCP | 47 TOOLS AVAILABLE |
|---|---|
| Code review agent | 4 tools, read only |
| Issue triage bot | 9 tools |
| Platform team assistant | all 47 tools |
- Tool by toolConnecting the server is not the unit of access. Each tool is granted on its own.
- Default denyA tool you have not enabled is never exposed to the agent.
- Approval on writesTools that change or delete things can require a human to confirm.
Is Todoist MCP secure?
One endpoint instead of raw credentials. Every call is checked against policy and logged.
| AI agentsConnect through your MCP client configuration. | NeuralTrustApply controls based on your configuration. | Todoist MCP47 documented tools, taken from the tool-name registry in Doist's own repository. Every tool carries readOnlyHint, destructiveHint and idempotentHint annotations, and all six update tools are annotated destructive because they replace values rather than merging them. |
|---|---|---|
| Coding assistants | Connection & credentialsper user | Tasks8 tools |
| Internal copilots | Tool access policiesdeny by default | General & Objects7 tools |
| Workflow automations | Request & response inspectioninline | Projects5 tools |
| Compatible MCP clients | Activity monitoringalerts and logs | All other tools27 tools |
Use Todoist MCP with any AI agent framework
Setup guides for the tools your team already uses.
Claude CodeAdd Todoist MCP to Claude Code from the terminal
CursorUse Todoist MCP tools inside Cursor's agent
CodexConnect Todoist MCP to the OpenAI Codex CLI
WindsurfGive Windsurf's Cascade access to Todoist MCP
Gemini CLICall Todoist MCP from Gemini CLI prompts
Google AntigravityLet Antigravity agents work with Todoist MCP
GitHub CopilotUse Todoist MCP in Copilot agent mode
Claude EnterpriseRoll out Todoist MCP across your Claude workspace
Microsoft Copilot StudioAdd Todoist MCP as a tool for Copilot Studio agents
Open WebUIBring Todoist MCP tools into Open WebUI chats
LangChainLoad Todoist MCP tools into LangChain agents
HaystackPlug Todoist MCP into Haystack agent pipelines
Strands AgentsGive Strands agents Todoist MCP tools
Vercel AI SDKCall Todoist MCP from Vercel AI SDK apps
n8nUse Todoist MCP in n8n AI agent workflows
Common questions about running Todoist MCP through TrustGate.
Other MCP servers
More servers from the catalog.
Calendly
Securely connect your AI agents to Calendly through TrustGate: event types, scheduled events, invitees, availability, scheduling links, routing forms and organization membership, scoped per agent and logged.
36 toolsViewAirtable
Securely connect your AI agents to Airtable through TrustGate: bases, tables, schemas, records, fields, comments and attachments, scoped per agent and logged.
40 toolsView)
Smartsheet
Securely connect your AI agents to Smartsheet through TrustGate: sheets, rows, columns, reports, dashboards, charts, discussions, attachments and automated workflows, scoped per agent and logged.
83 toolsViewFront
Securely connect your AI agents to Front through TrustGate: conversation search and timelines, drafts and sending, comments, tasks, tags, contacts and workspace structure, scoped per agent and logged.
28 toolsView
Connect Todoist MCP instantly
Pick the tools, point your agent at TrustGate, and Todoist MCP is live: scoped, inspected, logged.