rag-reference-discovery-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| USER_AGENT | No | User agent string used for scholarly API requests; see .env.example in the repository for details. | |
| ZOTERO_API_KEY | No | API key for optional Zotero Web API integration. | |
| ZOTERO_LIBRARY_ID | No | Zotero library ID corresponding to the supplied API key and library type. | |
| ZOTERO_LIBRARY_TYPE | No | Zotero library type (e.g. user or group) corresponding to the supplied API key. | |
| ZOTERO_LOCAL_API_URL | No | Optional URL for a local Zotero API endpoint; enables read-only checks against a local Zotero library. |
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 |
|---|---|
| analyze_research_textC | Analyze attachment-derived text into explainable scholarly search terms. |
| analyze_attachment_resourceC | Read a client-provided MCP resource when a standards-compliant resource adapter is available. |
| discover_referencesB | Discover and rank public scholarly metadata without claiming validation. |
| validate_referenceC | Independently validate a DOI using public metadata registries. |
| discover_and_validate_referencesC | Analyze supplied RAG text, discover candidates, and validate the strongest matches. |
| format_validated_referenceC | Format only metadata that has an accepted validation status. |
| healthA | Report non-secret server and integration configuration. |
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 map to distinct workflow stages, but analyze_research_text and analyze_attachment_resource share a similar 'analyze' prefix and discover_references vs discover_and_validate_references could cause some selection confusion. The descriptions mostly clear this up, so the overlap is minor.
The set largely follows a clear verb_noun snake_case pattern, e.g. discover_references, validate_reference, format_validated_reference. 'health' is a bare noun and 'discover_and_validate_references' uses a compound verb, creating minor deviations from the dominant pattern.
Seven tools is a well-scoped size for a reference discovery and validation server. Each tool supports a distinct part of the analyze-discover-validate-format workflow without unnecessary bloat.
The tool surface covers the full intended pipeline: extracting search terms, discovering references, validating DOIs, combining discovery and validation, and formatting results. The health tool also covers operational needs, and no critical workflow appears missing.