TokenHub MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| EXA_API_KEY | No | Selects Exa as the default search provider when Brave is not set. | |
| TAVILY_API_KEY | No | Selects Tavily as the default search provider when Brave and Exa are not set. | |
| SERPAPI_API_KEY | No | Selects SerpAPI as the default search provider when the other keyed providers are not set. | |
| BRAVE_SEARCH_API_KEY | No | Selects Brave as the default web_search provider and authenticates Brave Search. | |
| TOKENHUB_ENABLE_FS_MUTATIONS | No | When set to 'true', enables trusted-local filesystem write, move, and delete workflows. Leave unset for read/list behavior. | |
| TOKENHUB_ALLOW_PRIVATE_NETWORK | No | When set to 'true', allows trusted-local web and browser retrieval of localhost, private LAN, and other non-public network targets. Leave unset for public-network-only retrieval. |
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
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| discover_capabilitiesA | Find compact internal developer capabilities without loading their full schemas. |
| run_workflowD | Batch a common developer workflow server-side and return summaries plus resource links. |
| retrieve_contextC | Token-budgeted retrieval across files, Git state, and web pages. |
| read_resourceB | Progressively read a TokenHub resource by snippet, line range, or full content. |
| capture_stateC | Store logs, snapshots, or state summaries as resource-linked artifacts. |
| estimate_costB | Estimate tool cost and token savings before an expensive call. |
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 targets a distinct operation: discovery, workflow execution, state capture, context retrieval, resource reading, and cost estimation. The only slight potential confusion is between retrieve_context and read_resource, but their descriptions clearly separate broad retrieval from specific resource access.
All tool names follow a consistent verb_noun pattern with snake_case (discover_capabilities, run_workflow, capture_state, retrieve_context, read_resource, estimate_cost). This uniformity makes the tool set predictable and easy to navigate.
With 6 tools, the server is well-scoped for a context and resource management purpose. Each tool serves a clear function without redundancy, fitting comfortably within the ideal 3-15 tool range.
The set covers core operations for managing context and resources: discovery, execution, saving, retrieval, reading, and cost estimation. Minor gaps include no explicit update/delete operation for saved states or resources, but these may not be essential given the server's focus.