Claude MarketplacesSubmit

Neo — AI/ML Engineering

neoairesearch/neo-mcp

Connects Claude and other MCP-compatible editors to Neo's remote AI/ML execution backend. You describe a training job or data pipeline in plain language, Neo runs it on its infrastructure, then a local daemon writes all outputs directly to your workspace. The MCP layer exposes tools for submitting tasks, polling status, retrieving logs, sending feedback when Neo asks clarifying questions, and pausing or stopping long-running jobs. Also includes integration management for storing third-party credentials like GitHub, HuggingFace, and AWS tokens locally so Neo tasks can authenticate without re-prompting. Install via pip or npm, drop your API key in the editor config, and offload compute-heavy workflows without leaving your coding environment.

GitHub stars: 2
tools: 12
auth: Secret needed
transport: STDIO
env vars: 3
last updated: Jun 8, 2026

Tools

12 tools

Public tool metadata for what this MCP can expose to an agent.

neo_submit_taskSubmit a task to Neo. Returns `thread_id` immediately.

Submit a task to Neo. Returns `thread_id` immediately.

No parameter schema in public metadata yet.

neo_list_tasksList all running and recent tasks — useful after reopening your editor.

List all running and recent tasks — useful after reopening your editor.

No parameter schema in public metadata yet.

neo_task_statusCheck status: `RUNNING` / `COMPLETED` / `WAITING_FOR_FEEDBACK` / `PAUSED` / `TERMINATED`.

Check status: `RUNNING` / `COMPLETED` / `WAITING_FOR_FEEDBACK` / `PAUSED` / `TERMINATED`.

No parameter schema in public metadata yet.

neo_get_messagesRead full task output once status is `COMPLETED`. Capped at ~20 000 tokens.

Read full task output once status is `COMPLETED`. Capped at ~20 000 tokens.

No parameter schema in public metadata yet.

neo_send_feedbackReply to Neo when it asks a clarifying question (`WAITING_FOR_FEEDBACK`).

Reply to Neo when it asks a clarifying question (`WAITING_FOR_FEEDBACK`).

No parameter schema in public metadata yet.

neo_pause_taskPause a running task. Can be resumed.

Pause a running task. Can be resumed.

No parameter schema in public metadata yet.

neo_resume_taskResume a paused task.

Resume a paused task.

No parameter schema in public metadata yet.

neo_stop_taskPermanently stop and clean up a task.

Permanently stop and clean up a task.

No parameter schema in public metadata yet.

neo_list_integrationsList stored third-party API keys (names only — never the value).

List stored third-party API keys (names only — never the value).

No parameter schema in public metadata yet.

neo_add_integrationRegister a credential (GitHub, HuggingFace, Anthropic, OpenRouter, OpenAI, AWS S3, Weights & Biases, Kaggle) so Neo tasks can use it as an env var.

Register a credential (GitHub, HuggingFace, Anthropic, OpenRouter, OpenAI, AWS S3, Weights & Biases, Kaggle) so Neo tasks can use it as an env var.

No parameter schema in public metadata yet.

neo_test_integrationCall the provider's API to confirm a stored key is still valid.

Call the provider's API to confirm a stored key is still valid.

No parameter schema in public metadata yet.

neo_remove_integrationDelete a stored key from this machine.

Delete a stored key from this machine.

No parameter schema in public metadata yet.

Sponsors

Promote your product hereReach 125k+ visitors monthly Promote your product hereReach 125k+ visitors monthly

Agent CaffeineSponsoredClose your lid, come back to finished work

SnugSponsoredThe clipboard manager for the AI era Promote your product hereReach 125k+ visitors monthly

WaxSponsoredBlock anything with supervision, without erasing your iPhone Promote your product hereReach 125k+ visitors monthly

knobsSponsoredSee your site on any phone, any day, in any language

Configuration

NEO_SECRET_KEY*secret

Your Neo API secret key (starts with sk-v1-). Get one at https://heyneo.com.

NEO_ENVIRONMENTdefault: production

Neo backend environment: 'production' (default) or 'staging'.

NEO_WORKSPACE_DIR

Project root where Neo tasks read and write files. Must be the git/project root.

Readme

Neo MCP

Connect your AI editor to Neo's remote execution backend. Describe an AI/ML task in plain language — Neo trains the model, builds the pipeline, or runs the workload on its backend, then writes all output files directly to your local machine.

Works with Claude Code, Cursor, Windsurf, VS Code (GitHub Copilot), Zed, Continue.dev, and OpenAI Codex CLI.

Get your API key at heyneo.com/dashboard → Settings → API Keys.

How it works

Your editor  ──MCP──▶  neo-mcp server  ──API──▶  Neo backend
                            │                          │
                            └──────────────────▶  Local daemon
                                                  (writes files,
                                                   runs scripts)

You describe a task: "Train a fraud detection model on data.csv"
The editor calls neo_submit_task via MCP
Neo's backend processes the task and sends commands to the local daemon
The daemon runs on your machine — writing files, running scripts, installing packages
Output files appear directly in your workspace

View the full README on GitHub

Details

Repository: github.com/neoairesearch/neo-mcp
Website: heyneo.com
Package: neo-mcp
Transport: STDIO
Auth: Required
Registry: MCP registry, active