Task Hub MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| task_initA | Create a new task and start tracking it. Use this when the user wants to begin a new tracked task. The task will be saved to ~/.task-hub/tasks//. Args: name: Task name (will be slugified for the directory name) description: What this task is about tags: Optional tags for filtering (e.g. ["bioinfo", "ngs"]) |
| task_saveA | Save the current state of a task. Call this when the user says "save", "保存", or wants to checkpoint progress. The context parameter should be a concise summary of the current state — goals, what's done, what's next, key decisions. Args: name: Task name or ID context: Current context summary (distilled, not raw conversation) progress: Progress update to append decisions: New key decisions made (list of strings) pitfalls: New pitfalls discovered (list of strings) artifacts: Files to snapshot [{"path": "/abs/path", "description": "..."}] agent: Which agent is saving (e.g. "claude-code", "codex", "hermes") messages: Optional conversation messages to append to conversation.jsonl expected_revision: Revision returned by task_load; rejects stale saves |
| task_loadA | Load a task's context to continue working on it. Call this when the user says "load task X", "加载 X 任务", or wants to resume a specific task. Returns the distilled context (context.md) that gives you everything you need to continue the task. Args: name: Task name or ID include_conversation: If true, also include the full conversation log |
| task_listA | List all saved tasks. Call this when the user says "what tasks do I have", "有哪些任务", or wants to see what's available before loading one. Args: status: Filter by status ("active", "paused", "archived") tag: Filter by tag |
| task_resumeA | Continue a task, reactivating it when paused. Call this when the user says "continue where I left off", "继续上次的", or "resume". With no name, the most recent active task is loaded; if none is active, the most recent paused task is reactivated. Args: name: Optional task name or ID to load or reactivate |
| task_stopA | Stop tracking a task by pausing or archiving it. Call this when the user says "stop tracking", "不用追踪了", or wants to archive/finish a task. Args: name: Task name or ID reason: Why the task is being stopped archive: Permanently archive instead of pausing expected_revision: Revision returned by task_load; rejects stale updates |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 6 tools
Each tool targets a distinct lifecycle step: init creates, save checkpoints, list queries, load restores, resume reactivates, and stop pauses/archives. The only potential overlap between task_load and task_resume is resolved by task_resume's explicit role in reactivating paused tasks, while task_load requires a name and returns context.
All tools follow the consistent pattern task_<verb> in snake_case, with clear verbs (init, save, list, load, resume, stop). This makes the toolset predictable and easy to navigate.
With six tools, the server is well-scoped for task tracking: creation, state saving, listing, loading, resuming, and stopping. No redundancy or unnecessary tools are present.
The task lifecycle is fully covered: init creates a task, save persists progress, load retrieves context, resume reactivates, stop pauses or archives, and list provides an overview. No obvious missing operations for the stated purpose.