shelvd-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| SHELVD_DIR | No | Set SHELVD_DIR to use another directory instead of ~/.shelvd for storing ideas. |
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": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| park_ideaC | Park an idea for later, with an optional due date (natural language ok), project and context. |
| list_ideasC | List parked ideas as a markdown table. Show the table to the user exactly as returned; do not reformat or drop dates and ids. |
| get_ideaB | Get one idea with full context, notes and history. |
| complete_ideaC | Mark an idea done. |
| snooze_ideaC | Move an idea's due date later. Give |
| delete_ideaA | Permanently delete ONE idea by exact id. Confirm with the user first. Never guess an id from a vague title; list ideas to find it. |
| clear_ideasA | Delete many ideas: scope 'done', 'all' or 'project:'. Without confirm=true it only previews. Show the preview to the user and call again with confirm=true only after they approve. |
| update_ideaA | Edit an idea: title, details, project, due date, or append a |
| search_ideasB | Search ideas by text in title, details, project and notes. Shows a markdown table; show it as returned. |
| start_ideaC | Load an idea's details, context and notes as a brief so you can start work on it. |
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 10 tools
Most tools have clearly distinct purposes: create, retrieve, update, complete, snooze, delete, bulk-clear, and search are all separable. The main mild overlap is between get_idea and start_idea, since both retrieve an idea's context, though start_idea is framed as a workflow brief. list_ideas vs search_ideas is also close but still distinguishable by whether the user wants all ideas or a text query.
All tools follow a consistent snake_case verb_noun pattern: park_idea, list_ideas, get_idea, complete_idea, start_idea, snooze_idea, delete_idea, clear_ideas, update_idea, search_ideas. The naming is predictable and easy to scan.
Ten tools is well-scoped for an idea-parking workflow: core CRUD, lifecycle actions, bulk cleanup, and search are all represented without obvious padding. Each tool maps to a meaningful operation.
The surface covers the full idea lifecycle: create, list, get, update, start, complete, snooze, delete, bulk delete, and search. Minor gaps include no explicit reopen/uncomplete operation and no restore for deleted ideas, but most workflows can be handled with the existing tools.