reading-cottage
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| agent_id | No | Identifier for the agent. Used to keep the agent's notes separate from the human's. Defaults to 'agent'. | |
| data_dir | No | Path to the data directory where books, configuration, and reading records are stored. Defaults to ~/.reading-cottage. | |
| agent_name | No | Display name for the agent. Defaults to 'Reading Agent'. |
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 |
|---|---|
| reader_list_booksC | List the local bookshelf and shared reading progress. |
| reader_start_readingA | Return one chapter plus this agent's journal, pinned context, and previous discussion. This is context, not an unread-reply queue. Keep comments short and avoid later-chapter spoilers by convention; the tool does not enforce a spoiler access boundary. |
| reader_get_chapterA | Get a chapter with all paragraph annotations and highlights. |
| reader_add_annotationB | Write a paragraph annotation as this MCP server's stable agent identity. |
| reader_mark_chapter_doneB | Mark a chapter done and advance the book's shared progress. |
| reader_update_journalA | Append to this agent identity's separate per-book running journal. |
| reader_get_paragraph_threadA | Get one paragraph and its complete annotation thread. |
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 7 tools
Most tools are clearly distinct: listing books, annotating, marking done, updating the journal, and fetching paragraph threads all have separate purposes. However, reader_start_reading and reader_get_chapter both return chapter content, so an agent could be uncertain which to use if it simply wants to read a chapter.
All tools share the reader_ prefix and mostly follow a verb_noun pattern such as list_books, get_chapter, add_annotation, and update_journal. reader_start_reading and reader_mark_chapter_done deviate slightly from the pure noun-object form but are still predictable and readable.
Seven tools is a well-scoped count for a reading, annotation, progress, and journal workflow. Each tool serves a distinct purpose and the set does not feel bloated or thin.
The set covers the core lifecycle: list books, read chapters, annotate paragraphs, view annotation threads, advance progress, and update a journal. Minor gaps exist, such as no standalone journal retrieval or annotation update/delete, but reader_start_reading surfaces the journal and annotations appear designed as append-only.