toad-mcp-server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| HOST | No | HTTP transport bind address | 127.0.0.1 |
| PORT | No | HTTP transport port | 3100 |
| TRANSPORT | No | Transport mode: stdio or http | stdio |
| EVAL_FRAMEWORK_URL | No | Eval Framework endpoint | http://localhost:3002 |
| SEMANTIC_SEARCH_URL | No | Semantic Search Engine endpoint | http://localhost:3001 |
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": true
} |
| resources | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| toad_search_documentsA | Search documents using semantic similarity. Connects to the Semantic Search Engine to find relevant documents based on a natural language query. |
| toad_run_evalA | Run an evaluation suite from the Eval Framework. Executes test cases against a prompt variant and returns scores and pass/fail results. |
| toad_system_statusA | Check health of all portfolio services: Semantic Search API and Eval Framework. Returns status, latency, and error details for each component. |
| toad_list_promptsA | List available prompts from the Prompt Registry. Supports pagination and optional filtering by tag. |
| toad_get_promptA | Get a specific prompt by name with full metadata including version, score history, and template content. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| system-status | Current health status of all portfolio services |
| document-qa | Answer questions based on retrieved document context |
| code-review | Review code changes and suggest improvements |
| summarize-thread | Summarize a conversation thread into key points |
| eval-judge | LLM-as-judge prompt for evaluating model outputs |
| entity-extraction | Extract structured entities from unstructured text |
TDQS
Scored across 5 tools
Each tool targets a distinct capability: semantic search, eval execution, system health, prompt listing, and prompt detail retrieval. The relationship between list_prompts and get_prompt is complementary rather than ambiguous.
All tools follow the same `toad_` prefix with a clear verb_noun pattern: search_documents, run_eval, system_status, list_prompts, get_prompt. This makes the toolset highly predictable.
Five tools is well-scoped for a specialized server covering search, evaluation, system health, and prompt registry access. Each tool earns its place without redundant or unnecessary entries.
The server supports read/search/run workflows, but lacks prompt registry write operations (create/update/delete) and any way to list available eval suites. Agents may encounter dead ends when trying to discover eval inputs or manage prompt lifecycle.