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., "@genpark-instinct-proactive-intent-reflex-skill Infer my intent from recent cues and dry-run the next action with guardrails."
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.
🌐 GenPark MCP Hub • 📦 GenPark Official • 📖 Documentation
genpark-instinct-proactive-intent-reflex-skill is a deterministic, zero-dependency Python skill and native Model Context Protocol (MCP) server engineered for next-generation personal AI agents. It distills core architectural principles from Meta (ambient multimodal perception), Muse (continuous episodic memory), Instinct (zero-prompt proactive agency), and Jev (System-1 sub-millisecond typed decision cognition).
Executive Capability: Instinct AI-inspired zero-prompt proactive personal agent detecting user attention cues, computing intent probabilities, and executing guarded dry-run actions.
🐍 Zero External pip Dependencies: Runs instantaneously on standard Python 3.9+ with zero environment bloat.
🔌 Native Model Context Protocol (MCP): Plugs directly into Claude Desktop, Cursor IDE, Windsurf, and custom agent swarms.
🧠 System-1 Low-Latency Cognition: Slashes unnecessary frontier LLM invocations by routing routine and reflexive decisions at up to 200x faster execution speed.
🛡️ Safety & Privacy Guardrails: Enforces reversible execution checkpoints, strict token budgets, and local-first memory retention.
Related MCP server: genpark-instinct-proactive-intent-reflex-skill
graph LR
A[👁️ Ambient Perception: Meta / Screen] --> B[🧠 Instinct Proactive Sensor]
B --> C{⚡ Jev System-1 Decision Layer}
C -->|Fast Reflex / Cached Tool| D[🛠️ Deterministic Action]
C -->|Ambiguous / Multi-Hop Plan| E[🤔 System-2 Frontier LLM]
D --> F[(📜 Muse Episodic Memory Stream)]
E --> F
F -->|Decayed Context Briefing| Apython example_usage.pyfrom client import InstinctProactiveIntentReflex
client = InstinctProactiveIntentReflex()
result = client.run_benchmark_proactive_reflex()
print(result)Connect this skill to Claude Desktop, Cursor, or any MCP-compliant client:
claude_desktop_config.json{
"mcpServers": {
"genpark-instinct-proactive-intent-reflex-skill": {
"command": "python",
"args": ["/path/to/genpark-instinct-proactive-intent-reflex-skill/mcp_server.py"]
}
}
}python mcp_server.py --testParameter | Type | Required | Description |
|
| Yes | Primary context, state vector, or action candidate |
|
| Yes | Standardized schema containing typed decision outputs and telemetry |
Rather than sending every small interaction to heavy reasoning LLMs, this architecture implements Jev-style System-1 cognition and Instinct proactive sensing to execute fast, deterministic, schema-enforced routing and guardrails.
Memory records utilize Muse-style Ebbinghaus forgetting curves with recency decay, contradiction resolution, and user-controlled deletion cascades.
This server cannot be deployed
Zero-secret MCP gateway for AI agents: risk-scored, audited calls with human-in-the-loop approval.
Persistent, verbatim, searchable memory for AI assistants — one memory across every MCP client.
Private, portable memory and reusable skills for AI agents.
Persistent personal memory for AI assistants — save, search, and recall across every MCP client.
