Skip to content

Integration Guide

clinvk is designed to integrate seamlessly with AI development workflows, from Claude Code Skills to LangChain agents to CI/CD pipelines.

Integration Overview

clinvk integrates with various tools and frameworks through its HTTP API:

Integration Methods:

Category Tools API Endpoint
AI Development Claude Code Skills, LangChain/LangGraph, Custom Agents /api/v1/* or /openai/v1/*
SDKs OpenAI SDK, Anthropic SDK /openai/v1/* or /anthropic/v1/*
Automation CI/CD Pipelines, Shell Scripts /api/v1/*

Data Flow:

┌─────────────────┐     ┌─────────────────┐     ┌─────────────────┐
  AI Development            SDKs               Automation      ─────────────         ────────────          ────────────     Claude Skills          OpenAI SDK           CI/CD Pipelines│
  LangChain      │────▶│   Anthropic SDK │────▶│  Shell Scripts    Custom Agents       └─────────────────┘     └─────────────────┘
└─────────────────┘                                                                      └───────────┬───────────┘
                                                                                ┌─────────────────┐
                                     clinvk server                                     └────────┬────────┘
                                                                   ┌────────────────────┼────────────────────┐
                                                                              ┌──────────────┐   ┌──────────────┐   ┌──────────────┐
                 Claude CLI       Codex CLI        Gemini CLI                 └──────────────┘   └──────────────┘   └──────────────┘

Integration Methods

Method Use Case API Endpoint
Claude Code Skills Extend Claude with multi-backend capabilities /api/v1/*
LangChain/LangGraph AI framework integration /openai/v1/*
CI/CD Automated code review, documentation /api/v1/*
Client Libraries Python, TypeScript, Go clients All endpoints
MCP Server Model Context Protocol integration Available

Quick Integration Examples

Claude Code Skill

# In your skill script
curl -s http://localhost:8080/api/v1/prompt \
  -H "Content-Type: application/json" \
  -d '{"backend": "gemini", "prompt": "Analyze this data"}'

LangChain

from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    base_url="http://localhost:8080/openai/v1",
    model="claude",
    api_key="not-needed"
)

OpenAI SDK

from openai import OpenAI

client = OpenAI(
    base_url="http://localhost:8080/openai/v1",
    api_key="not-needed"
)

GitHub Actions

- name: AI Code Review
  run: |
    payload=$(jq -n --arg prompt "Review:\n${{ steps.diff.outputs.changes }}" '{backend:"claude", prompt:$prompt}')
    curl -sS -X POST http://localhost:8080/api/v1/prompt \
      -H "Content-Type: application/json" \
      -d "$payload"

Prerequisites

Before integrating, ensure:

  1. clinvk is installed: See Installation
  2. Backend CLIs are available: At least one of claude, codex, or gemini
  3. Server is running: clinvk serve --port 8080

Choosing the Right Endpoint

Your Situation Recommended Endpoint
Using OpenAI SDK or LangChain /openai/v1/*
Using Anthropic SDK /anthropic/v1/*
Building Claude Code Skills /api/v1/*
Need parallel/chain execution /api/v1/*
Simple REST integration /api/v1/*

Next Steps

Choose your integration path: