Task Orchard
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| TASK_ORCHARD_DB | No | Path to the SQLite database file. | ~/.local/share/task-orchard/orchard.db |
| TASK_ORCHARD_TZ | No | Your timezone as an IANA name, such as `Europe/London`. | The system timezone, else UTC. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
Task Orchard keeps one person's tasks as trees and helps them decide what to do with their time and energy. It stores and computes facts; you make the judgements: what matters, how to break work down, what to suggest.
How to work with it:
At the start of a conversation, call get_guide with name "overview" unless you have already read it. It explains the concepts and lists the other guides.
Before a multi-step job (triage, breaking a task down, planning a block of time, closing a task), call get_guide for that job and follow it.
Task ids are numbers that come from tool results. Never invent one.
Times in results are already in the user's timezone and come with a relative phrase such as "in 2 days". Use those; do not do date arithmetic.
If a tool returns an error, read its hint and correct the call.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| logging | {} |
| prompts | {
"listChanged": true
} |
| resources | {
"subscribe": false,
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| get_guideA | Read one of Task Orchard's guides: step-by-step instructions for a job. Call with no name to list the guides. Call with a name to get its text, then follow it. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| overview | Read this first. What Task Orchard is, the ideas behind it, and which guide to use for which job. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 1 tool
With only one tool, there is no possibility of misselection or overlapping purpose. The tool's role is unambiguous.
The single name get_guide follows a clear verb_noun snake_case convention. Only one data point exists, so consistency across the set cannot be fully demonstrated.
One tool is thin for a server whose domain is browsing and consuming multiple guides. It is workable but feels under-scoped for even a read-only surface.
Listing and fetching guide text are covered, but there is no search, categorization, or metadata operation to help an agent find the right guide among many. For a read-only guide service this is notable but not fatal.