Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@calibre-mcp list all books in the Foundation series"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP server for Calibre book library management. Designed for use with Claude Code.
Talks to your local Calibre library via calibredb CLI — no Content Server needed.
Tool | Description |
| Search books using Calibre's query syntax ( |
| Get full metadata for a book by ID |
| Get absolute EPUB file path (for downstream processing) |
| List all books in a series, sorted by index |
| List custom columns defined in the library |
| Set or append custom column values for a book |
| Edit standard metadata fields (title, authors, tags, series, etc.) |
Related MCP server: calibre-manager
claude mcp add calibre -- uvx --from git+https://github.com/xmkevinchen/calibre-mcp.git calibre-mcpAdd -s user to make it available across all projects:
claude mcp add -s user calibre -- uvx --from git+https://github.com/xmkevinchen/calibre-mcp.git calibre-mcpOr add to your project .mcp.json (project-level only):
{
"mcpServers": {
"calibre": {
"command": "uvx",
"args": ["--from", "git+https://github.com/xmkevinchen/calibre-mcp.git", "calibre-mcp"],
"env": {
"CALIBRE_LIBRARY_PATH": "/path/to/Calibre Library"
}
}
}
}claude mcp add calibre -- uvx --from /path/to/calibre-mcp calibre-mcpVariable | Description | Default |
| Path to Calibre library directory |
|
| Path to | auto-detected via PATH |
Search for a book:
search_books("title:pirates past noon")Get EPUB path for reading:
get_epub_path(247) # returns /Users/you/Calibre Library/Author/Book (247)/book.epubBrowse a series:
list_series("Magic Tree House")uvx caches installed packages. To pull the latest version:
uvx --reinstall --from git+https://github.com/xmkevinchen/calibre-mcp.git calibre-mcpThen restart Claude Code to pick up the changes.
calibredb not found — Make sure Calibre is installed and calibredb is in your PATH. On macOS, Calibre installs CLI tools to /Applications/calibre.app/Contents/MacOS/. Add it to PATH or set CALIBREDB_PATH.
claude mcp add doesn't work — Use add-json instead:
claude mcp add-json -s user calibre '{"command":"uvx","args":["--from","git+https://github.com/xmkevinchen/calibre-mcp.git","calibre-mcp"],"env":{"CALIBRE_LIBRARY_PATH":"/path/to/Calibre Library"}}'Or create .mcp.json manually in your project root (see above).
Search and retrieval (search_books, get_book_info, get_epub_path, list_series)
Custom columns and metadata (get_custom_columns, set_custom_column, set_metadata)
MIT
Get detailed metadata for a book by its Calibre ID.
Returns title, authors, series, tags, formats, file paths, and all other metadata.
Args: book_id: The Calibre book ID
| Name | Required | Description | Default |
|---|---|---|---|
| book_id | Yes |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description bears full burden. It discloses the tool returns metadata fields (title, authors, etc.) but does not mention side effects, auth requirements, or error conditions (e.g., invalid book_id). Acceptable but not thorough.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is extremely concise: two sentences for the main description and a docstring-style parameter listing. Every part adds value, no fluff.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool has only one parameter, schema coverage 0%, and an output schema exists (reducing need to explain return values), the description covers the essential behavior. Minor omission: no mention of what happens if book_id not found, but acceptable for simplicity.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 0%, so the description must explain parameters. It clearly defines book_id as 'The Calibre book ID', adding semantic meaning beyond the schema type 'integer'. No other parameters exist, making this sufficient.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool retrieves detailed metadata for a book by its Calibre ID, specifying the exact resource (book) and action (get info). It distinguishes from siblings like search_books (search) and set_metadata (update).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies the tool is for retrieving full metadata when you know the book_id, but does not explicitly contrast with siblings or mention when not to use it (e.g., for searching by title, use search_books).
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List all custom columns defined in the Calibre library.
Custom columns are user-defined metadata fields beyond the standard fields (title, author, tags, etc.).
Args: details: Include column types, default values, and other metadata
| Name | Required | Description | Default |
|---|---|---|---|
| details | No |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations provided, so description carries full burden. It correctly portrays a read-only list operation, and mentions the optional 'details' parameter effect. Could add that it returns all columns with no filters or pagination.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Very concise three sentences: one states purpose, one contrasts with standard fields, one describes parameter. No wasted words.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given 1 parameter and output schema exists, description covers the input well but doesn't hint at output structure or potential size. Still adequate for a simple list tool.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 0%, so description must compensate. It explains the only parameter 'details' with a clear purpose: 'include column types, default values, and other metadata'. This adds meaning beyond the schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Description clearly states the tool lists custom columns, distinguishes from standard fields, and contrasts with siblings like set_custom_column and get_book_info.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Description implies use when you need to see all custom columns but lacks explicit when-not or alternatives. Sibling names hint at related tools but not mentioned.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Get the absolute file path of a book's EPUB file.
This is the core tool for the reading-quiz workflow — returns the EPUB path so it can be unzipped and chapters read directly.
Args: book_id: The Calibre book ID
| Name | Required | Description | Default |
|---|---|---|---|
| book_id | Yes |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Without annotations, the description carries the burden. It indicates the tool provides an absolute file path and that the result is used for unzipping. It does not specify side effects, but given its read-only nature inferred from 'get', this is adequate. No contradictions with annotations (none provided).
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is very concise: two sentences plus an args section. Each part serves a purpose: stating the action, contextualizing within a workflow, and documenting the parameter. No unnecessary words.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given it has an output schema (not shown, but present), the description does not need to explain return values. It explains the parameter sufficiently for a simple tool. The only missing aspect is potentially clarifying that the path is absolute, which it does. It is complete for its complexity.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
With 0% schema description coverage and only one parameter, the description adds value by explaining the parameter's purpose ('book_id: The Calibre book ID') in the args section, but this is minimal. The schema already provides the parameter name and type. The description does not elaborate on format or constraints.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool returns the absolute file path of a book's EPUB file. It also specifies the purpose within the reading-quiz workflow, distinguishing it from siblings by emphasizing it's for obtaining the path to enable unzipping and reading.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description explains the core purpose for a specific workflow, implying usage when needing the EPUB path. However, it does not explicitly mention when not to use it or provide alternatives among siblings. The requirement for a book_id is clear from the schema.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List all books in a series, sorted by series_index.
Great for seeing the full reading order of a series like "Magic Tree House".
Args: series_name: Name of the series to look up fields: Fields to include in results
| Name | Required | Description | Default |
|---|---|---|---|
| series_name | Yes | ||
| fields | No |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Fully describes the sorting behavior (by series_index) and the purpose. No annotations provided, so the description carries full burden. It doesn't mention any destructive or side effects (none expected). Could mention that only top-level books are listed or if nested series are included, but overall good.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Short and clear, front-loaded with main purpose. The example adds value but slightly increases length. Could be more concise if example removed, but it helps understanding.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given no annotations and an output schema exists, the description suffices for a straightforward list operation. Doesn't explain return format (but output schema covers that). No coverage for what constitutes a 'series' or error handling, but acceptable for a simple tool.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 0% (parameters not described in schema), so description must compensate. The description briefly mentions 'series_name' and 'fields' in Args, but adds no detail beyond what the parameter names imply. For example, it doesn't explain what 'fields' values are valid beyond being a list of strings.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Clearly states it lists all books in a series sorted by series_index, which is a specific verb+resource+sorting detail. Distinguishes from siblings like 'get_book_info' which retrieves a single book or 'search_books' which filters by criteria.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Describes when to use (to see reading order) with a concrete example. However, doesn't explicitly say when not to use or suggest alternatives like 'search_books' or 'get_book_info'.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Search books in the Calibre library using Calibre's search syntax.
Search syntax examples:
Simple text: "robot" (searches all fields)
Field-specific: author:asimov, title:"magic tree house", tag:fiction
Boolean: author:asimov AND title:foundation
Series: series:"magic tree house"
Args: query: Search expression using Calibre's query language fields: Fields to include in results limit: Max number of results (omit for all) sort_by: Field to sort by
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | ||
| fields | No | ||
| limit | No | ||
| sort_by | No | title |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries full burden for behavioral disclosure. It does not mention read vs. write behavior, potential performance impacts, or what happens with invalid queries. The description focuses on syntax examples but lacks transparency on side effects or limitations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is front-loaded with the tool's purpose, then provides structured examples and parameter explanations. It is concise but each section adds value. Minor improvement: the 'Args' section repeats parameter info partially redundant with schema.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the complexity (search syntax) and that an output schema exists (reducing need to describe return values), the description covers query usage well. However, it lacks information about error handling or behavior when no results match.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The description explains the query parameter with syntax examples, adds meaning for fields, limit, and sort_by beyond the schema (which has no descriptions). With 0% schema coverage, the description compensates well for most parameters, though it could clarify acceptable values for fields and sort_by.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states that the tool searches books using Calibre's search syntax. It specifies the resource (books) and the action (search), and distinguishes from siblings like get_book_info which retrieves info for a specific book.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides examples of search syntax and parameters, implicitly guiding usage. However, it does not explicitly state when to use this tool versus alternatives (e.g., get_book_info for specific books), nor does it mention any prerequisites or caveats.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Set the value of a custom column for a book.
Custom columns are referenced by their lookup name (e.g. "#status", "#rating"). Use get_custom_columns() to see available columns.
Args: book_id: The Calibre book ID column: Custom column lookup name (e.g. "#status") value: Value to set append: If True, append to existing values instead of replacing
| Name | Required | Description | Default |
|---|---|---|---|
| book_id | Yes | ||
| column | Yes | ||
| value | Yes | ||
| append | No |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description discloses behavior: it sets values, supports appending via the 'append' parameter. Since no annotations are provided, the description carries the full burden of behavioral disclosure. It explains the effect of the 'append' parameter well, but does not mention any destructive implications or error conditions beyond standard input validation.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is concise and well-structured: a one-line summary, a usage tip, then a clear parameter list. Every sentence adds value, and there is no extraneous information.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's moderate complexity, annotations are missing, and schema coverage is 0%, the description provides adequate parameter explanations and a cross-reference to get_custom_columns. It could mention return value (since output schema exists, not required) but the description is largely complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 0%, so the description must compensate. It defines each parameter: book_id (Calibre book ID), column (lookup name format with example), value (value to set), and append (explanation of behavior). This adds significant meaning beyond the schema's type-only definitions.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool sets the value of a custom column for a book, specifying the action ('set'), resource ('custom column'), and scope ('for a book'). It distinguishes from siblings by referencing get_custom_columns() for lookup.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides explicit guidance on how to reference columns via lookup name and directs to get_custom_columns() for available columns. It implies usage context but does not explicitly state when not to use this tool or provide alternative approaches beyond listing siblings.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Set standard metadata fields for a book.
Supported fields: title, authors, author_sort, comments, cover, isbn, languages, pubdate, publisher, rating, series, series_index, sort, tags, title_sort, identifiers.
Note: this REPLACES field values. For tags, pass all desired tags comma-separated. For identifiers: "isbn:123456,goodreads:789".
Args: book_id: The Calibre book ID fields: Dict of field_name → value, e.g. {"tags": "sci-fi,classics", "series": "Foundation"}
| Name | Required | Description | Default |
|---|---|---|---|
| book_id | Yes | ||
| fields | Yes |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Clearly states that field values are REPLACED (behavioral trait). Provides details on how tags and identifiers should be formatted. Since no annotations exist, description handles transparency well, though it could mention input validation limits.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Description is well-structured with a brief intro, a bullet list of supported fields, important notes, and an Args section. It is relatively concise but the list of fields could be more compact.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given no annotations, description covers purpose, supported fields, replacement behavior, and format requirements. Output schema exists (context signal) so return values are not needed. Slight lack of guidance on error handling or prerequisite conditions.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 0%, so description must compensate. It explains book_id is 'the Calibre book ID' and fields is a dict mapping field names to values with examples. This adds significant meaning beyond schema property names.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Description specifies verb 'set' and resource 'metadata fields for a book' with a clear list of supported fields. It distinguishes from siblings like 'get_book_info' and 'set_custom_column' by focusing on standard metadata.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Description explains when to use (setting standard metadata) and provides usage examples for tags and identifiers. However, it does not explicitly state when not to use or mention alternatives like set_custom_column for custom fields.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v0.1.0get_book_infoget_custom_columnsget_epub_pathlist_seriessearch_booksset_custom_columnset_metadataScored across 7 tools
Each tool has a clearly distinct purpose: reading vs writing, metadata vs custom columns, search vs series listing. No overlap.
All tool names use verb_noun pattern (get_, list_, search_, set_) consistently, with clear imperative verbs.
7 tools is well-scoped for a Calibre library server, covering search, read, and write operations without bloat.
Covers search, read metadata, get file path, list series, set metadata, manage custom columns. Missing delete or create book operations, but core workflows are covered.
MCP server for Russian books search, details, and recommendation candidates.
Read-only MCP server exposing a user ORANO library to their own AI agent.
An MCP server that gives your AI access to the source code and docs of all public github repos
