A professional-grade integration skill for building, debugging, and optimizing Kimi API workflows with a focus on reliability and operational safety.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install kimi
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install kimi 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 API Workflow Integration?
The Kimi skill for Openclaw Skills provides a robust framework for developers to integrate Moonshot AI's powerful LLM capabilities into their local environments and agentic workflows. It excels at handling live model verification, request shaping, and migration from other OpenAI-compatible providers, ensuring that your AI implementations are both stable and cost-effective.
By prioritizing technical truth over cached data, this skill ensures you are always using live model IDs and optimized request patterns. It creates a structured local memory environment to manage workload routing, sensitive-data redaction, and experimental prompt testing, making it an essential tool for developers building high-stakes AI applications.
Kimi API Workflow Integration Use Cases
- Deploying Kimi for long-context research and complex reasoning tasks.
- Migrating existing OpenAI-compatible codebases to the Moonshot API without breaking changes.
- Implementing deterministic JSON outputs for downstream tool-calling and automation.
- Managing cost and performance by routing specific workloads to the most efficient Kimi models.
- Debugging API authentication, retries, and request-response patterns in real-time.
How Kimi API Workflow Integration Works
- Discovery: The skill fetches live model IDs directly from the Moonshot API to ensure connectivity and valid routing.
- Classification: It categorizes the user request into specific workloads like fast chat, coding, or long-context research to apply the best prompt strategy.
- Sanitization: Before sending data, it applies redaction rules from the local approvals schema to protect sensitive information.
- Execution: It dispatches the request to the official Moonshot endpoint using optimized request patterns and failure recovery logic.
- Persistence: With user approval, it saves successful routes, debug logs, and stateful preferences to the local filesystem for future consistency.
Kimi API Workflow Integration Setup
To get started with Kimi via Openclaw Skills, ensure you have your API key ready and run the following commands:
# Install the skill
clawhub install kimi
# Set your environment variable
export MOONSHOT_API_KEY="your_key_here"
# Initialize the memory structure
mkdir -p ~/kimi/
Kimi API Workflow Integration Data Schema & Taxonomy
The skill maintains a structured local directory at ~/kimi/ to manage workflow state and security boundaries:
| File | Purpose |
|---|---|
memory.md |
Tracks activation rules, status, and stable default models. |
routes.md |
Maps specific workload types to their preferred Kimi model IDs. |
approvals.md |
Defines redaction preferences and sensitive data boundaries. |
experiments.md |
Stores notes on prompt performance and fallback strategies. |
logs/ |
Contains sanitized payloads for troubleshooting recurring issues. |
Kimi API Workflow Integration Advanced Features
- Live Model Verification: Automatically polls
api.moonshot.aito prevent errors caused by stale or decommissioned model IDs. - OpenAI Migration Playbook: Dedicated logic for mapping OpenAI request shapes to Kimi-specific requirements.
- Dual-Pass Logic: Supports a second normalization pass for workflows requiring high-precision structured JSON output.
- Stateless-First Architecture: Operates without local footprints unless explicitly configured by the user for persistent workflows.
- Multi-Skill Integration: Seamlessly connects with
coding,backend, andfastapiskills to build full-stack AI services.
SKILL.md
Loading
Related Openclaw Skills
DataVault
A local-first CLI tool for scanning, classifying, and reporting on data assets across 60+ file formats.
HTML Report Generator
A precision-engineered tool that converts any input into a series of structured, presentation-quality HTML report pages.
Summarize
A high-performance CLI tool designed to distill URLs, local files, and multimedia content into concise summaries using leading AI models.
A-Stock Daily Report
An automated financial reporting tool that fetches and analyzes A-share market performance, including indices and trending sectors.
Blackbox CLI
A sophisticated multi-model coding agent that executes tasks across several LLMs and utilizes an internal judge to select the highest quality code implementation.
EdrawMax AI Diagram Generator
A professional AI-driven skill that transforms natural language descriptions into high-quality visual diagrams and source code.
Featured*
