opencode-document-rag-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
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_documentsA | List supported documents under DOCS while excluding DOCS/mdDB. |
| ingest_documentA | Convert one DOCS document to Markdown with Marker and index its structural chunks. |
| ingest_all_documentsC | Synchronize supported DOCS documents, excluding the mdDB output directory. |
| search_documentsC | Search ChromaDB and return complete Markdown sections containing nearby tables and LaTeX equations. |
| read_markdown_sectionB | Read one complete section from the stored Markdown by its breadcrumb path. |
| index_statusA | Show indexed documents, Markdown paths, and chunk counts. |
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 6 tools
Each tool has a clearly distinct purpose: listing documents, ingesting a single document, ingesting all documents, searching, reading a section, and checking index status. The only potential overlap (ingest_document vs. ingest_all_documents) is clearly resolved by singular vs. plural scope, leaving no ambiguity for an agent.
Most tools follow a verb_noun pattern (list_, ingest_, search_, read_), but 'index_status' deviates by starting with a noun. Despite this minor inconsistency, all names are lowercase, underscore-separated, and self-descriptive, so the pattern remains predictable.
With only 6 tools, the server is well-scoped for a document RAG system. Each tool addresses a distinct step in the workflow—discovery, ingestion (single and bulk), retrieval, reading, and status monitoring—without redundancy or excess.
The surface covers the core lifecycle: listing available documents, ingesting new ones (single or all), searching indexed content, reading specific sections, and checking index status. Missing delete/update operations are minor gaps since the server focuses on additive indexing and retrieval, which is acceptable for its purpose.