literature-evidence-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
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| search_documentsA | Search one verified snapshot. The default bm25 mode stays local and offline; explicit enhanced mode may send bounded query and candidate content through a separately configured transport. |
| get_excerptA | Return one bounded evidence excerpt after verifying its snapshot, document_id, and chunk_id association. |
| get_multiple_excerptsA | Return one to five bounded evidence excerpts in requested chunk_id order; all IDs must belong to the selected document and snapshot. |
| get_document_metadataA | Return path-free document and asset metadata from one verified local snapshot. |
| get_document_tocA | Return bounded navigation derived from indexed Markdown headings or PDF pages; it does not claim to be an author-provided table of contents. |
| read_document_sectionA | Read at most 1200 characters from a section_id previously returned for the same snapshot and document. |
| find_in_documentA | Search inside one document using the same local SQLite FTS5/BM25 expression and deterministic tie order as search_documents. |
| retrieval_statusA | List libraries, list one library's successful snapshots, or verify one explicit library_id and snapshot_id pair. |
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 8 tools
Each tool targets a distinct operation: corpus search, within-document search, single or multiple excerpt retrieval, TOC navigation, section reading, metadata lookup, and repository status. The closest pair, search_documents and find_in_document, is clearly separated by scope.
Most tools consistently follow a verb_noun snake_case pattern such as get_document_metadata, search_documents, and read_document_section. The only notable deviation is retrieval_status, which uses a noun phrase instead of a verb-first name.
Eight tools is well-scoped for a focused literature-evidence retrieval server. Each tool provides a clear, non-redundant capability needed for searching, navigating, and extracting verified evidence.
The tool set covers the core workflow from searching a snapshot to retrieving metadata, TOC, sections, and excerpts. The main gap is the lack of an explicit list_documents operation for browsing all documents in a snapshot, though search can discover documents.