TaskChampion MCP
@GabiUp2
About TaskChampion MCP
TaskChampion MCP Server
Config
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"taskchampion": {
"command": "uvx",
"args": [
"taskchampion-mcp"
]
}
}
}Tools
No tools detected
Fetch the live tool list by running this server in a temporary sandbox using the button above.
Overview
What is TaskChampion MCP?
TaskChampion MCP is a Model Context Protocol server that lets LLMs read, create, modify, and manage Taskwarrior 3.x tasks and Timewarrior time entries. It wraps the task and timew CLI tools and exposes them as structured MCP tools for any compatible AI assistant. It is for users who self-host TaskChampion and want LLMs to help with project planning, task decomposition, and time tracking.
How to use TaskChampion MCP?
Install via pip install taskchampion-mcp or uv tool install taskchampion-mcp, then configure your IDE’s MCP settings to use the command taskchampion-mcp-server with no arguments. On first run, the server boots in onboarding mode; you must set role and schema/schema_path in ~/.config/taskchampion-mcp/config.toml — either through an LLM-driven walkthrough, the CLI wizard (./dev.sh init), or by hand-editing the config.
Key features of TaskChampion MCP
- Three cumulative permission roles: CONTRIBUTOR, GENERATOR, MANAGER
- Bundled task schema presets (minimal, gtd, scrum, kanban, custom)
- No shell execution – subprocess argument lists only
- Input sanitization and configurable rate limiting
- Full audit logging with timestamp and result code
- Dry-run mode and confirmation tokens for destructive operations
- Sensitive field redaction in LLM responses
Use cases of TaskChampion MCP
- Let an LLM assist with project planning and task breakdown.
- Automate time tracking by having an AI create Timewarrior entries.
- Delegate task creation and modification while keeping full control.
- Manage a Taskchampion task database from LLM-powered editors (Neovim, Cursor, VS Code, Windsurf, Claude Desktop).
FAQ from TaskChampion MCP
What happens on first run?
On a fresh install with no config, the server boots in onboarding mode. All tools are visible in tools/list, but operational tools return structured schema_unset errors until you set role and schema/schema_path in ~/.config/taskchampion-mcp/config.toml.
How do I set up roles and schemas?
Set role (CONTRIBUTOR / GENERATOR / MANAGER) and schema or schema_path under the [server] section in config.toml. You can use the LLM-driven initialization process, the CLI wizard (./dev.sh init), or hand-edit the file and then call reload_configuration.
What permission levels does TaskChampion MCP support?
Three cumulative roles: CONTRIBUTOR (read, annotate, modify), GENERATOR (plus create), and MANAGER (plus complete and delete). Set the role in ~/.config/taskchampion-mcp/config.toml.
Why are tools returning errors after connecting?
If tools return schema_unset errors, ~/.config/taskchampion-mcp/config.toml is missing either role or schema/schema_path. Add both and call reload_configuration or restart the MCP server.
Is the server secure?
Yes. All CLI calls use subprocess argument lists (never shell=True), LLM inputs are sanitized against allowlists, configurable rate limiting caps requests per minute/hour, every operation is audit-logged, and destructive actions support dry-run previews and confirmation tokens.
Frequently asked questions
What happens on first run?
On a fresh install with no config, the server boots in onboarding mode. All tools are visible in `tools/list`, but operational tools return structured `schema_unset` errors until you set `role` and `schema`/`schema_path` in `~/.config/taskchampion-mcp/config.toml`.
How do I set up roles and schemas?
Set `role` (CONTRIBUTOR / GENERATOR / MANAGER) and `schema` or `schema_path` under the `[server]` section in `config.toml`. You can use the LLM-driven initialization process, the CLI wizard (`./dev.sh init`), or hand-edit the file and then call `reload_configuration`.
What permission levels does TaskChampion MCP support?
Three cumulative roles: CONTRIBUTOR (read, annotate, modify), GENERATOR (plus create), and MANAGER (plus complete and delete). Set the role in `~/.config/taskchampion-mcp/config.toml`.
Why are tools returning errors after connecting?
If tools return `schema_unset` errors, `~/.config/taskchampion-mcp/config.toml` is missing either `role` or `schema`/`schema_path`. Add both and call `reload_configuration` or restart the MCP server.
Is the server secure?
Yes. All CLI calls use subprocess argument lists (never `shell=True`), LLM inputs are sanitized against allowlists, configurable rate limiting caps requests per minute/hour, every operation is audit-logged, and destructive actions support dry-run previews and confirmation tokens.
Basic information
More Other MCP servers
MCP Go 🚀
mark3labsA Go implementation of the Model Context Protocol (MCP), enabling seamless integration between LLM applications and external data sources and tools.
XcodeBuildMCP
cameroncookeA Model Context Protocol (MCP) server and CLI that provides tools for agent use when working on iOS and macOS projects.
AutoBrowser MCP
autobrowser-aiBrowser MCP is a Model Context Provider (MCP) server that allows AI applications to control your browser

Figs, Relationship Expert
Figs O'Sullivan, LMFTCouples therapist Figs O'Sullivan (LMFT, 17 years, endorsed by Sue Johnson, creator of EFT) inside your AI. Ask relationship questions answered from his clinical writing, name the pattern in a fight, get a three-step rep
Mcp
browsermcpBrowser MCP is a Model Context Provider (MCP) server that allows AI applications to control your browser
Comments