Overview

flowchart LR
    A[Channels] -->|user message| B[OpenClaw Gateway]
    B -->|gemma4:e2b plans| C[agentic-rag skill]
    C -->|rag_query.sh POST /predict| D[LitServe API]
    D --> E[Researcher Agent]
    E --> F[Vector DB Tool]
    E --> G[Firecrawl Search]
    E --> H[Writer Agent]
    H --> I[JSON Response]
    F --> J[(Qdrant)]
    G --> K[Firecrawl API]
    B --> L[(Ollama gemma4:e2b)]
    E --> L
    H --> L
    I --> C
    C --> B
    B -->|reply| A

1. User messages OpenClaw on Telegram, WhatsApp, or CLI
2. gemma4:e2b handles chat and may invoke the agentic-rag skill
3. Skill runs rag_query.sh → LitServe POST /predict on port 8001
4. CrewAI Researcher + Writer use Qdrant (and optional Firecrawl)
5. Answer returns through OpenClaw to the channel

Layer	Role
OpenClaw	Channels, sessions, tools, skills, daemon
gemma4:e2b	Fast local chat + tool planning (~7GB)
agentic-rag skill	Calls your LitServe /predict endpoint
qwen-agentic-rag	Two-agent RAG API (same gemma4:e2b via .env)

{ width="100%" }

1. RAG API (terminal A)

cd guides/qwen-agentic-rag
python -m venv .venv && source .venv/bin/activate
pip install -r requirements.txt
ollama pull gemma4:e2b
cp ../openclaw-gemma-rag/env.rag.example .env   # OLLAMA_MODEL=ollama/gemma4:e2b
python setup_vectordb.py   # once
python server.py           # default http://127.0.0.1:8001

2. OpenClaw + Gemma (terminal B)

cd guides/openclaw-gemma-rag && source ./use-node22.sh   # Node 22+ required
ollama pull gemma4:e2b
npm install -g openclaw@latest
openclaw onboard --install-daemon
openclaw models set ollama/gemma4:e2b

Merge config/openclaw.snippet.json5 (in this guide folder) into ~/.openclaw/openclaw.json, then:

cd guides/openclaw-gemma-rag
chmod +x install-skill.sh skills/agentic-rag/scripts/*.sh
./install-skill.sh
openclaw gateway restart

3. Test

# RAG health
RAG_API_URL=http://127.0.0.1:8001 ./skills/agentic-rag/scripts/rag_health.sh

# OpenClaw agent (uses gemma; may call RAG skill for ML questions)
openclaw agent --message "What is cross-validation? Use the knowledge base if helpful." --thinking low

Project layout

Path	Purpose
skills/agentic-rag/SKILL.md	OpenClaw skill instructions
skills/agentic-rag/scripts/rag_query.sh	POST query to LitServe
install-skill.sh	Copy skill into ~/.openclaw/workspace/skills/
use-node22.sh / .nvmrc	Switch to Node 22+ for OpenClaw CLI
env.rag.example	Gemma .env for the RAG crew
test-local.sh	Smoke test: Ollama, RAG API, skill scripts
assets/openclaw-gemma-rag-workflow.gif	Animated architecture diagram
assets/render_workflow_gif.py	Regenerate GIF from HTML source
config/openclaw.snippet.json5	Model + skill env sample config
TUTORIAL.md	Full setup (channels, security, troubleshooting)

Read the full integration tutorial →{ .md-button .md-button--primary }