LaunchKit · 2026
Back to Skills

aetherlang-karpathy-skill

Implement 10 advanced AI agent node types for any DSL/runtime system — plan compiler, code interpreter, critique.

2
442 downloads
by @contrario

Setup & Installation

openclaw skills install @contrario/aetherlang-karpathy-skill

Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:

npx clawhub install aetherlang-karpathy-skill

Version History

v1.0.3Mar 10, 2026 - Version note by skill developer:

v1.0.3: Karpathy-inspired, operator_note added

What This Skill Does

API connector for the AetherLang Omega service at api.neurodoc.app. Executes 10 agent node types including planning, code interpretation, routing, ensemble synthesis, and parallel processing. All computation runs server-side with no local execution or credentials required.

Combines multi-step agent orchestration, sandboxed code execution, and parallel API calls through a single public endpoint, removing the need to wire up individual services or manage a local agent runtime.

When to use it

  • Run sandboxed Python math without a local runtime
  • Fetch live crypto prices and summarize in one flow
  • Generate content with multiple AI personas then synthesize
  • Loop a prompt over a list of items like cuisines or cities
  • Quality-gate an AI response by retrying until a score threshold is met

Example Workflow

Here's how your AI assistant might use this skill in practice.

INPUT

User asks: What is the current Bitcoin price and what does it mean for the market?

AGENT
  1. 1Build a flow with a tool node pointing to the CoinGecko price API
  2. 2Chain a transform node to extract and reformat the JSON response
  3. 3Pass the result to an LLM node for market commentary
  4. 4Send the assembled flow code and query as a POST request to api.neurodoc.app/aetherlang/execute
  5. 5Parse result.outputs.result.response from the JSON response
OUTPUT

Current Bitcoin price with a brief market analysis paragraph