This MCP server exposes a ComputeFlux/OpenAI-compatible endpoint to MCP hosts so agents can list models and run chat completions as tools.
List available model IDs with computeflux_models.
Run a non-streaming chat completion with computeflux_chat using required model and OpenAI-style messages.
Optionally set max_tokens (default 512) and temperature (default 1.0).
Receive assistant text plus metadata: model, finish reason, latency, and usage.
Point it at any OpenAI-compatible API via COMPUTEFLUX_BASE_URL; default is ComputeFlux.
Connect from any MCP client/host over stdio JSON-RPC, including Cursor, Claude Desktop, and Windsurf.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@mcp-computeflux list the models available on ComputeFlux"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Zero-dependency MCP (Model Context Protocol) server. Give any MCP host — Cursor, Claude Desktop, Windsurf, any MCP client — one OpenAI-compatible endpoint, and call models on ComputeFlux as plain tools: list what's available, then run a chat completion.
Your MCP agents hit flash-tier models over one OpenAI-compatible endpoint.
~175 lines of Node 18+, no npm install to run the server (only the optional
client/ e2e harness pulls SDK deps).
Backend-agnostic: point it at any OpenAI-compatible API — ComputeFlux is the default, not the only target.
ComputeFlux is an OpenAI-compatible, TEE-verifiable multi-model inference gateway deployed on the Polkadot testnet.
Tool | What it does |
| Lists the model ids served by the endpoint. |
| Runs a non-streaming chat completion ( |
Related MCP server: mcp-llm-gateway
Variable | Required | Default | Notes |
| yes | — | Bearer key for the endpoint. |
| no |
| Any OpenAI-compatible base URL. |
export COMPUTEFLUX_API_KEY="your_c***_key"
# export COMPUTEFLUX_BASE_URL="https://your-openai-compat-host/v1" # optionalPublished on the official MCP Registry (.mcpb bundle, no npm step):
io.github.computeflux2026isgod/mcp-computeflux — listed on the official MCP Registry (search by this name).
Any MCP client / agent that supports registry install (mcp add, Cursor, Claude Desktop,
Claude Code) can pull it by name. Direct git-clone quickstart:
git clone https://github.com/computeflux2026isgod/mcp-computeflux
export COMPUTEFLUX_API_KEY="your_...key"
node /path/to/mcp-computeflux/server.mjs # stdio JSON-RPCmcpServers snippet:
{
"mcpServers": {
"computeflux": {
"command": "node",
"args": ["/path/to/mcp-computeflux/server.mjs"],
"env": { "COMPUTEFLUX_API_KEY": "your_c***_key" }
}
}
}client/ contains a tiny harness using the official MCP SDK — the exact path
Cursor / Claude Desktop use. It spawns the server over stdio, lists tools, and runs
one chat completion with a unique-token round-trip fidelity check.
cd client && npm install && npm run e2e # expects COMPUTEFLUX_API_KEY in the .env it points toNewline-delimited JSON-RPC 2.0 over stdio. Only protocol messages are written to stdout; logs go to stderr.
MIT
Send a chat completion request to the ComputeFlux (OpenAI-compatible) endpoint. Returns the assistant message text plus usage. Use model like "CAO/deepseek-flash".
| Name | Required | Description | Default |
|---|---|---|---|
| model | Yes | Model id, e.g. "CAO/deepseek-flash" | |
| messages | Yes | Chat messages (OpenAI schema). | |
| max_tokens | No | Max tokens to generate (optional). | |
| temperature | No | Sampling temperature (optional, default 1.0). |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full burden. It does disclose useful behavior beyond the schema: the endpoint is OpenAI-compatible and the call returns the assistant message text plus usage. However, it omits auth requirements, error behavior, rate limits, and whether responses stream, leaving notable gaps for a network call.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two short sentences, front-loaded with the core action. The return-value sentence earns its place since there is no output schema, but the trailing model example is redundant with the schema description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Return values are covered, which compensates for the absent output schema, and all parameters are documented in the schema. Still, for a remote chat-completion call with no annotations, the description says nothing about authentication, failure modes, or streaming, so it is only minimally complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the schema already documents all four parameters, establishing a baseline of 3. The description's model example ('CAO/deepseek-flash') merely duplicates the schema's own example and adds no new semantic detail.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description names a specific verb and resource ('Send a chat completion request to the ComputeFlux endpoint'), which is immediately distinguishable from the sibling computeflux_models. It does not explicitly name or contrast with that sibling, but the action is unambiguous.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
There is no guidance on when to use this tool versus alternatives, no prerequisites, and no exclusions. The only usage-adjacent text is the model-format hint, which is really a parameter note rather than selection guidance.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List the models available on the ComputeFlux (OpenAI-compatible) endpoint.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full disclosure burden, and it says almost nothing behavioral. It does not confirm the operation is read-only/non-mutating, note authentication or endpoint-configuration requirements, or hint at rate limits or pagination. 'List' implies a read, but that is inference rather than disclosure.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
A single, front-loaded sentence that names the action, the resource, and the endpoint. Every word earns its place with no redundancy or filler.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a zero-parameter, read-only discovery tool this is close to adequate, but with no output schema the description could reasonably state what the list returns (e.g., model identifiers) and whether authentication is required. Those gaps keep it at a minimum-viable level.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The schema declares zero parameters at 100% coverage, so the baseline of 4 applies and there is nothing for the description to document. No misleading parameter language is present.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific verb (List) and resource (models) scoped to the ComputeFlux endpoint, which an agent can act on immediately. It does not explicitly contrast itself with the sibling computeflux_chat, but the list-vs-chat distinction is self-evident from the resource named.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Usage is only implied: an agent can infer this is for discovering which models the endpoint exposes before issuing a chat request. There is no explicit statement of when to call it or how it relates to computeflux_chat, and no caveats about ordering or eligibility.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v1.0.0computeflux_chatcomputeflux_modelsScored across 2 tools
The two tools have clearly distinct purposes: listing available models versus sending a chat completion request. There is no overlap in functionality or ambiguous selection risk.
Both tools share the same 'computeflux_' prefix and use snake_case, which is consistent. The only minor deviation is that one name is a plural noun and the other is a verb-like action, but the pattern remains readable.
Two tools is slightly under the typical 3–15 range, but for a narrow OpenAI-compatible wrapper covering model listing and chat completions, each tool earns its place. The surface is minimal but reasonable for the stated purpose.
Core chat functionality and model listing are present, but an OpenAI-compatible endpoint often includes other common operations such as embeddings, completions, or audio. These gaps are notable, though agents could still perform basic chat workflows.
MCP server for OpenAI API (chat completions, image generation, embeddings) via AceDataCloud
Hosted MCP server for LLM cost estimation, model comparison, and budget-aware routing.
MCP server that lets AI assistants use all OneSchema features exposed via the public API.
An MCP server that provides an API to LLMs to manage their JumpCloud resources.
