← Back to Skills Marketplace
67
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install open-ragflow
Description
RAGFlow open-source Retrieval-Augmented Generation (RAG) engine — deployment, configuration, management, and troubleshooting.
README (SKILL.md)
\r \r
Open RAGFlow\r
\r Open-source RAG engine fusing RAG with Agent capabilities. Full-stack: Python backend (Flask), React/TypeScript frontend, Docker-deployed microservices.\r \r
Use Cases\r
\r
- Deploying or self-hosting RAGFlow via Docker Compose or from source\r
- Configuring RAGFlow (LLM providers, API keys, document engines, ports)\r
- Managing knowledge bases, datasets, documents, agents, and chats via RAGFlow CLI\r
- Understanding RAGFlow architecture (DeepDoc, Agent system, RAG pipeline)\r
- Integrating with RAGFlow REST API\r
- Troubleshooting RAGFlow deployment or runtime issues.\r \r \r
Quick Reference\r
\r
- Website: https://ragflow.io\r
- Cloud: https://cloud.ragflow.io\r
- Docs: https://ragflow.io/docs/dev/\r
- Repo: https://github.com/infiniflow/ragflow\r
- Discord: https://discord.gg/NjYzJD3GM3\r
- Docker Hub:
infiniflow/ragflow\r - License: Apache 2.0\r \r
When to Use Which Reference\r
\r
- Deploying / troubleshooting deployment → references/deployment.md\r
- Understanding architecture / components / data flow → references/architecture.md\r
- Using CLI to manage datasets, agents, models → references/cli-reference.md\r \r
Prerequisites\r
\r
- CPU >= 4 cores, RAM >= 16 GB, Disk >= 50 GB\r
- Docker >= 24.0.0 & Docker Compose >= v2.26.1\r
vm.max_map_count>= 262144 (Linux, for Elasticsearch)\r- gVisor: optional, only needed for the code executor (sandbox) feature\r \r
Docker Deployment (Quick Start)\r
\r
git clone https://github.com/infiniflow/ragflow.git\r
cd ragflow/docker\r
docker compose -f docker-compose.yml up -d\r
docker logs -f docker-ragflow-cpu-1 # wait for the banner, then login\r
# Open http://YOUR_SERVER_IP in browser\r
```\r
\r
Configure LLM API keys in `docker/service_conf.yaml.template` under `user_default_llm`, then restart:\r
\r
```bash\r
docker compose -f docker-compose.yml up -d\r
```\r
\r
### Chinese Mirror Images\r
\r
If Docker Hub is slow:\r
- Huawei Cloud: `swr.cn-north-4.myhuaweicloud.com/infiniflow/ragflow`\r
- Alibaba Cloud: `registry.cn-hangzhou.aliyuncs.com/infiniflow/ragflow`\r
\r
Set HuggingFace mirror if needed: `HF_ENDPOINT=https://hf-mirror.com` in `docker/.env`.\r
\r
## Key Configuration Files\r
\r
| File | Scope |\r
|------|-------|\r
| `docker/.env` | Environment variables: `SVR_HTTP_PORT`, `MYSQL_PASSWORD`, `MINIO_PASSWORD`, `DOC_ENGINE`, `RAGFLOW_IMAGE`, `HF_ENDPOINT` |\r
| `docker/service_conf.yaml.template` | Backend services: LLM factory, API keys, embedding/rerank/ASR/TTS models |\r
| `docker/docker-compose.yml` | Full stack orchestration |\r
| `docker/docker-compose-base.yml` | Infrastructure services only (dev mode) |\r
\r
## Post-Deployment LLM Setup\r
\r
1. Log in via browser\r
2. Edit `docker/service_conf.yaml.template`:\r
```yaml\r
user_default_llm:\r
factory: "OpenAI" # or "DeepSeek", "Gemini", etc.\r
api_key: "sk-..."\r
base_url: "https://api.openai.com/v1/"\r
```\r
3. Run `docker compose -f docker-compose.yml up -d` to apply.\r
\r
## CLI Quick Reference\r
\r
All CLI commands end with `;`. Full reference: [references/cli-reference.md](references/cli-reference.md).\r
\r
```bash\r
# Datasets\r
LIST DATASETS;\r
CREATE DATASET 'my_kb' WITH EMBEDDING 'text-embedding-ada-002' PARSER 'pdf';\r
DROP DATASET 'my_kb';\r
LIST FILES OF DATASET 'my_kb';\r
\r
# Documents\r
IMPORT '/path/to/doc.pdf' INTO DATASET 'my_kb';\r
PARSE DATASET 'my_kb' SYNC;\r
PARSE DATASET 'my_kb' ASYNC;\r
\r
# Search\r
SEARCH 'What is RAG?' ON DATASETS 'my_kb';\r
\r
# Models\r
CREATE MODEL PROVIDER 'openai' 'sk-...';\r
SET DEFAULT LLM 'gpt-4';\r
LIST MODEL PROVIDERS;\r
LIST DEFAULT MODELS;\r
\r
# Agents & Chats\r
LIST AGENTS;\r
LIST CHATS;\r
CREATE CHAT 'my_session';\r
DROP CHAT 'my_session';\r
\r
# Connection\r
PING;\r
SHOW CURRENT USER;\r
```\r
\r
## Switching Doc Engine (Elasticsearch → Infinity)\r
\r
```bash\r
docker compose -f docker/docker-compose.yml down -v # WARNING: clears data\r
# Edit docker/.env: set DOC_ENGINE=infinity\r
docker compose -f docker-compose.yml up -d\r
```\r
\r
Infinity is lighter weight but Linux/arm64 is not officially supported.\r
\r
## Architecture at a Glance\r
\r
```\r
Web UI (React+TS+vitejs+shadcn) → Flask API (/api/) → RAG Core (/rag/) + Agent (/agent/)\r
↓\r
Infrastructure: MySQL + Elasticsearch/Infinity + Redis + MinIO\r
```\r
\r
- **Backend** (`/api/`): Flask blueprints — kb, dialog, document, canvas, file, user\r
- **RAG Core** (`/rag/`): DeepDoc parsing, LLM/embedding/rerank abstractions, chunking, GraphRAG\r
- **Agent** (`/agent/`): Canvas-based workflow builder with components (LLM, Retrieval, Code Executor, MCP, Search, SQL)\r
- **Frontend** (`/web/`): React 18 + TypeScript + Vite\r
\r
See [references/architecture.md](references/architecture.md) for detailed component breakdown.\r
\r
## Development from Source\r
\r
```bash\r
git clone https://github.com/infiniflow/ragflow.git && cd ragflow\r
uv sync --python 3.12 && uv run python3 download_deps.py\r
docker compose -f docker/docker-compose-base.yml up -d\r
# Add to /etc/hosts: 127.0.0.1 es01 infinity mysql minio redis sandbox-executor-manager\r
source .venv/bin/activate && export PYTHONPATH=$(pwd)\r
bash docker/launch_backend_service.sh\r
# Separate terminal:\r
cd web && npm install && npm run dev\r
```\r
\r
## Troubleshooting Quick Reference\r
\r
| Problem | Fix |\r
|---------|-----|\r
| `network abnormal` browser error | Wait for Docker logs to show the RAGFlow banner — server is initializing |\r
| Docker pull timeout in China | Use `RAGFLOW_IMAGE` mirrors (Huawei Cloud / Alibaba Cloud) |\r
| HuggingFace unreachable | `export HF_ENDPOINT=https://hf-mirror.com` |\r
| ARM64 platform | Build Docker image from source (no official ARM64 image) |\r
| Port conflict | Change `80:80` to `\x3CPORT>:80` in `docker-compose.yml` |\r
| Elasticsearch exits with 137 | Increase Docker memory allocation |\r
| `vm.max_map_count` too low | `sudo sysctl -w vm.max_map_count=262144` |\r
\r
## API & SDK\r
\r
- REST API: `http://SERVER_IP/api/` — Swagger docs at `/api/docs`\r
- Python SDK: available in `sdk/python/`\r
- CLI client: `python admin/client/ragflow_cli.py \x3Ccommand>`\r
- Guide: [https://ragflow.io/docs/dev/](https://ragflow.io/docs/dev/)
Usage Guidance
Install or use this skill only if you intend to deploy or administer RAGFlow. Verify the upstream repo/images, review every Docker or CLI command before running it, back up data before destructive operations, protect API keys and default service passwords, restrict public network exposure, and import only documents you are comfortable storing and indexing.
Capability Analysis
Type: OpenClaw Skill
Name: open-ragflow
Version: 1.0.0
The skill bundle provides legitimate documentation and instructions for deploying and managing the RAGFlow open-source RAG engine. It includes standard Docker deployment procedures, CLI references for managing datasets and model providers, and architectural overviews. While the instructions involve high-privilege actions such as Docker orchestration and system configuration (e.g., modifying vm.max_map_count for Elasticsearch), these are strictly aligned with the stated purpose of the tool. No indicators of malicious intent, data exfiltration, or unauthorized persistence were found across the SKILL.md or reference files.
Capability Tags
Capability Assessment
Purpose & Capability
The purpose and capabilities align: the artifacts describe deploying, configuring, managing, and troubleshooting RAGFlow. The included Docker, CLI, document-ingestion, and agent-system capabilities are expected for that purpose but can affect local services and data.
Instruction Scope
The commands are presented as user-run setup/reference examples, not automatic execution. Some commands can delete RAGFlow data or change deployment state, so they should be run only after explicit user confirmation.
Install Mechanism
There is no install spec and no bundled executable code; setup relies on cloning the upstream RAGFlow repository, Docker images, and package managers. Users should verify the upstream source and versions before running them.
Credentials
Docker services, exposed ports, high resource requirements, LLM API keys, and stored/indexed documents are proportionate for self-hosting a RAG system, but users should secure networking, credentials, and imported data.
Persistence & Privilege
The guide starts daemonized Docker services and includes host-level setup such as sysctl and /etc/hosts edits. This persistence and privilege are disclosed and purpose-aligned for deployment.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install open-ragflow - After installation, invoke the skill by name or use
/open-ragflow - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
- Initial release of open-ragflow: open-source RAG engine with agent capabilities.
- Full-stack deployment: Python (Flask) backend, React/TypeScript frontend, Docker-based microservices.
- Supports deployment via Docker Compose and self-hosting from source.
- CLI and REST API for managing knowledge bases, datasets, agents, and chats.
- Comprehensive documentation and troubleshooting included for deployment and configuration.
- Compatible with multiple LLM providers, embedding models, and document engines (Elasticsearch/Infinity).
Metadata
Frequently Asked Questions
What is RAGFlow open-source Retrieval-Augmented Generation (RAG) engine — deployment, configuration, management, and troubleshooting.?
RAGFlow open-source Retrieval-Augmented Generation (RAG) engine — deployment, configuration, management, and troubleshooting. It is an AI Agent Skill for Claude Code / OpenClaw, with 67 downloads so far.
How do I install RAGFlow open-source Retrieval-Augmented Generation (RAG) engine — deployment, configuration, management, and troubleshooting.?
Run "/install open-ragflow" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is RAGFlow open-source Retrieval-Augmented Generation (RAG) engine — deployment, configuration, management, and troubleshooting. free?
Yes, RAGFlow open-source Retrieval-Augmented Generation (RAG) engine — deployment, configuration, management, and troubleshooting. is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does RAGFlow open-source Retrieval-Augmented Generation (RAG) engine — deployment, configuration, management, and troubleshooting. support?
RAGFlow open-source Retrieval-Augmented Generation (RAG) engine — deployment, configuration, management, and troubleshooting. is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created RAGFlow open-source Retrieval-Augmented Generation (RAG) engine — deployment, configuration, management, and troubleshooting.?
It is built and maintained by OpenLark (@openlark); the current version is v1.0.0.
More Skills