Model Context Protocol (MCP) server for planning and executing tasks.
Taskflow MCP server exists for a simple reason — assistants are far more useful when they can act on Taskflow directly instead of describing what you should do. Model Context Protocol (MCP) server for planning and executing tasks.
Once Taskflow is connected, these are the calls the assistant has available:
Configuration — By default, TaskFlow MCP saves tasks to tasks.yaml in the current working directory. You can customize this by setting the TASK_MANAGER_FILE_PATHplan_task — Register a new user request and plan its associated tasks (with optional subtasks)get_next_task — The get_next_task tool exposed by this servermark_task_done — The mark_task_done tool exposed by this serveropen_task_details — The open_task_details tool exposed by this serverlist_requests — The list_requests tool exposed by this serveradd_tasks_to_request — The add_tasks_to_request tool exposed by this serverupdate_task — The update_task tool exposed by this serverdelete_task — The delete_task tool exposed by this serveradd_subtasks — The add_subtasks tool exposed by this servermark_subtask_done — The mark_subtask_done tool exposed by this serverupdate_subtask — The update_subtask tool exposed by this servertaskflow-mcp on npm is all you need. Most clients run it directly, so configuration is a few lines and a restart.
You will need 4 environment variables: TASK_MANAGER_FILE_PATH, TASK_MANAGER_BASE_DIR, ARCHIVE_FILE_PATH, ARCHIVE_MODE. Keep credentials in your client's env block or a secrets manager rather than in a file you might commit.
Plenty of file and storage access servers cover similar ground. The differences that matter in practice are scope of access and how much setup stands between you and a working tool call. Taskflow's toolset — Configuration, plan_task, get_next_task and 11 more — is a fair guide to whether it matches your workflow. It is maintained by sizzlebop; worth a glance at recent repository activity before you build anything load-bearing on it.
We check each listing at SyncDev against the project's documentation before it goes live — if something here drifts out of date, it is a bug worth reporting.
| Tool | What it does |
|---|---|
| Configuration | By default, TaskFlow MCP saves tasks to tasks.yaml in the current working directory. You can customize this by setting the TASK_MANAGER_FILE_PATH environment variable: |
| plan_task | Register a new user request and plan its associated tasks (with optional subtasks). |
| get_next_task | The get_next_task tool exposed by this server. |
| mark_task_done | The mark_task_done tool exposed by this server. |
| open_task_details | The open_task_details tool exposed by this server. |
| list_requests | The list_requests tool exposed by this server. |
| add_tasks_to_request | The add_tasks_to_request tool exposed by this server. |
| update_task | The update_task tool exposed by this server. |
| delete_task | The delete_task tool exposed by this server. |
| add_subtasks | The add_subtasks tool exposed by this server. |
| mark_subtask_done | The mark_subtask_done tool exposed by this server. |
| update_subtask | The update_subtask tool exposed by this server. |
| delete_subtask | The delete_subtask tool exposed by this server. |
| export_task_status | Export the current status of all tasks in a request to a file. It's recommended to use absolute paths for more reliable file creation. |
{
"mcpServers": {
"taskflow": {
"command": "npx",
"args": ["-y", "taskflow-mcp"],
"env": {
"TASK_MANAGER_FILE_PATH": "your-value",
"TASK_MANAGER_BASE_DIR": "your-value",
"ARCHIVE_FILE_PATH": "your-value",
"ARCHIVE_MODE": "your-value"
}
}
}
}Add to claude_desktop_config.json, then restart Claude Desktop.
| Variable | Description | Required |
|---|---|---|
| TASK_MANAGER_FILE_PATH | Filesystem location the server is allowed to use. | Optional |
| TASK_MANAGER_BASE_DIR | Filesystem location the server is allowed to use. | Optional |
| ARCHIVE_FILE_PATH | Filesystem location the server is allowed to use. | Optional |
| ARCHIVE_MODE | Configuration value read at startup. | Optional |
Scoped local file access — read, write, search and reorganise files in directories you explicitly allow.
Search and read your Drive — Docs, Sheets and files become context your assistant can actually use.
Query, modify and analyse local SQLite databases in conversation — the fastest way to chat with a data file.
Build a programmable telecommunications stack for connecting telephony services with the Internet via a cloud-based utility.
Chat with your second brain — search, read and write vault notes through the Local REST API.
Connects AI models to an Obsidian knowledge base for direct access and manipulation of notes and folders.