Kimi Search for Openclaw

A high-performance web search skill for Openclaw agents powered by the Moonshot Kimi API's native $web_search tool.

chrisbdz
v1.0.1
Mar 7, 2026
0
1.7k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install kimi-search

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install kimi-search using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Kimi Search?

Kimi Search is a technical integration designed for the Openclaw Skills ecosystem that enables AI agents to perform real-time web searches. By leveraging the Moonshot Kimi API and its specialized $web_search tool, this skill allows agents to transcend their training data cutoffs to retrieve current news, updated documentation, and live data in both Chinese and English.

Built on the robust kimi-k2.5 model, the skill supports a massive 256k context window, ensuring that even information-heavy search results are processed accurately. It is a critical tool for developers using Openclaw Skills who require their agents to have grounded, factual, and up-to-the-minute information from the internet.

Kimi Search Use Cases

  • Fetching the latest software release notes or API documentation updates during a coding session.
  • Researching real-time market trends or news to inform technical decision-making.
  • Verifying facts or gathering live data points for data analysis tasks.
  • Enhancing AI agent reasoning with contextually relevant search results from the live web.

How Kimi Search Works

  1. The AI agent or user invokes the kimi-search tool with a specific search query.
  2. The skill script authenticates with the Moonshot API using the securely stored MOONSHOT_API_KEY.
  3. A request is sent to the kimi-k2.5 model with the $web_search builtin tool enabled.
  4. Kimi performs the web search on its server-side and processes the raw web content.
  5. The model synthesizes the search findings into a detailed answer tailored to the user's query.
  6. The skill outputs a structured JSON response containing the synthesized answer and token usage metrics.

Kimi Search Setup

To get started with this skill in your Openclaw Skills environment, first install the required Python dependency:

pip3 install openai

Next, configure your API key in your ~/.openclaw/openclaw.json file:

{
  "skills": {
    "entries": {
      "kimi-search": {
        "enabled": true,
        "apiKey": "sk-your-moonshot-api-key"
      }
    }
  }
}

Finally, restart the gateway to enable the tool:

openclaw gateway restart

Kimi Search Data Schema & Taxonomy

The Kimi Search skill returns a structured JSON object to ensure compatibility with various agent workflows. The schema includes:

Property Type Description
query string The search keyword or phrase provided to the tool.
answer string The comprehensive answer synthesized from the search results.
usage object Contains prompt_tokens, completion_tokens, and total_tokens for cost tracking.

Kimi Search Advanced Features

  • Uses the high-capacity kimi-k2.5 model for superior reasoning and large context handling.
  • Native support for both Chinese and English queries for global research needs.
  • Secure environment variable injection via the Openclaw Skills architecture.
  • Detailed token usage reporting, accounting for both model processing and the $0.03 flat fee per search invocation.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Requires
Github Stars: 0
forks: 0

Featured*