RAGFlow

https://ragflow.io/docs
Admin
RAGFlow is an open-source RAG engine based on deep document understanding that provides truthful...
Tokens:326,784
Snippets:2,314
Trust Score:8.6
Update:2 weeks ago
Show doc for...
Tokens:
Raw
### Test sandbox setup

Source: https://ragflow.io/docs/sandbox_quickstart

Activates the virtual environment, sets PYTHONPATH, installs dependencies, and runs the sandbox security tests.

```bash
source .venv/bin/activate  
export PYTHONPATH=$(pwd)  
uv pip install -r executor_manager/requirements.txt  
uv run tests/sandbox_security_tests_full.py  
```

--------------------------------

### SHOW VERSION Example

Source: https://ragflow.io/docs/admin_cli

Example of showing the version.

```text
RAGFlow(admin)> SHOW VERSION;  
```

--------------------------------

### Start the frontend with npm run dev

Source: https://ragflow.io/docs/launch_ragflow_from_source

Run these commands in the web directory to install dependencies and start the frontend dev server. With API_PROXY_SCHEME=go, the frontend routes /api and /v1 requests to port 9380 and /api/v1/admin requests to port 9381. Open http://127.0.0.1:9222/ unless Vite prints a different port.

```bash
cd web  
npm install  
API_PROXY_SCHEME=go npm run dev  

```

--------------------------------

### HELP example

Source: https://ragflow.io/docs/admin_cli

Example invocation of the help command.

```sql
RAGFlow(admin)> \help
```

--------------------------------

### Start Admin with --init-superuser

Source: https://ragflow.io/docs/admin_service

Starts the Admin Service and initializes the default superuser. Use only when initial account setup is required. If no superuser exists, it creates `admin@ragflow.io` with initial password `admin`; change that password immediately after first login. It does not reset an existing superuser's password.

```bash
RAGFLOW_DEV_MODE=true ./bin/ragflow_server --admin --init-superuser  

```

--------------------------------

### Start API server with --api

Source: https://ragflow.io/docs/admin_service

Starts the Go API server in development mode. Start after the Admin Service so it can report heartbeats.

```bash
RAGFLOW_DEV_MODE=true ./bin/ragflow_server --api  

```

--------------------------------

### LIST CONFIGS Example

Source: https://ragflow.io/docs/admin_cli

Example invocation of the LIST CONFIGS command.

```sql
RAGFlow(admin)> LIST CONFIGS;
```

--------------------------------

### Install RAGFlow CLI on Linux/macOS (default)

Source: https://ragflow.io/docs/admin_cli

Installs the prebuilt Go CLI from the latest GitHub Release. The installer detects OS and CPU architecture, downloads the matching binary, verifies it against SHA256SUMS, and installs it to /usr/local/bin/ragflow-cli by default.

```bash
curl -fsSL https://raw.githubusercontent.com/infiniflow/ragflow/main/tools/scripts/install.sh | sh  

```

--------------------------------

### LIST ENVS Example

Source: https://ragflow.io/docs/admin_cli

Example invocation of the LIST ENVS command.

```sql
RAGFlow(admin)> LIST ENVS;
```

--------------------------------

### Get Recent Messages - usage example

Source: https://ragflow.io/docs/python_api_reference

Example usage of RAGFlow.get_recent_messages. Pass a list of memory IDs.

```python
from ragflow_sdk import RAGFlow  
rag_object = RAGFlow(api_key="<YOUR_API_KEY>", base_url="http://<YOUR_BASE_URL>:9380")  
rag_object.get_recent_messages(["your memory_id"])  
```

--------------------------------

### SHOW ADMIN SERVER Example

Source: https://ragflow.io/docs/admin_cli

Example of showing the Admin Service connection.

```text
RAGFlow(admin)> SHOW ADMIN SERVER;  
```

--------------------------------

### Get Message Content - usage example

Source: https://ragflow.io/docs/python_api_reference

Example usage of Memory.get_message_content. Requires a Memory object initialized with a memory ID.

```python
from ragflow_sdk import RAGFlow,Memory  
rag_object = RAGFlow(api_key="<YOUR_API_KEY>", base_url="http://<YOUR_BASE_URL>:9380")  
memory_object = Memory(rag_object, {"id": "your memory_id"})  
memory_object.get_message_content(message_id)  
```

--------------------------------

### SHOW CURRENT Example

Source: https://ragflow.io/docs/admin_cli

Example of showing current CLI state.

```text
RAGFlow(admin)> SHOW CURRENT;  
```

--------------------------------

### Get Memory Config - usage example

Source: https://ragflow.io/docs/python_api_reference

Example of retrieving the configuration of a memory. Requires a Memory object initialized with the memory ID.

```python
from ragflow_sdk import RAGFlow, Memory  
rag_object = RAGFlow(api_key="<YOUR_API_KEY>", base_url="http://<YOUR_BASE_URL>:9380")  
memory_object = Memory(rag_object, {"id": "your memory_id"})  
memory_object.get_config()  

```

--------------------------------

### Get agent by ID with RAGFlow.get_agent()

Source: https://ragflow.io/docs/python_api_reference

Gets a single agent by its ID and returns the detailed agent payload. The example shows how to call get_agent with a placeholder ID and print the returned Agent object.

```python
RAGFlow.get_agent(agent_id: str) -> Agent  
```

```python
from ragflow_sdk import RAGFlow  
  
rag_object = RAGFlow(api_key="<YOUR_API_KEY>", base_url="http://<YOUR_BASE_URL>:9380")  
agent = rag_object.get_agent("AGENT_ID")  
print(agent)  
```

--------------------------------

### Run all setup with Makefile

Source: https://ragflow.io/docs/sandbox_quickstart

Runs all setup, build, launch, and tests with a single command using the Makefile.

```bash
make  
```

--------------------------------

### Get Recent Messages - success response

Source: https://ragflow.io/docs/http_api_reference

Successful response for the Get Recent Messages endpoint. Returns a code of 0 and a data array containing message objects. The example shows one message and indicates that other messages are omitted.

```json
{  
    "code": 0,  
    "data": [  
        {  
            "agent_id": "8db9c8eddfcc11f0b5da84ba59bc53c7",  
            "content": "User Input: what is pineapple?\nAgent Response: A pineapple is a tropical fruit known for its sweet, tangy flavor and distinctive, spiky appearance. Here are the key facts:\nScientific Name: Ananas comosus\nPhysical Description: It has a tough, spiky, diamond-patterned outer skin (rind) that is usually green, yellow, or brownish. Inside, the juicy yellow flesh surrounds a fibrous core.\nGrowth: Unlike most fruits, pineapples do not grow on trees. They grow from a central stem as a composite fruit, meaning they are formed from many individual berries that fuse together around the core. They grow on a short, leafy plant close to the ground.\nUses: Pineapples are eaten fresh, cooked, grilled, juiced, or canned. They are a popular ingredient in desserts, fruit salads, savory dishes (like pizzas or ham glazes), smoothies, and cocktails.\nNutrition: They are a good source of Vitamin C, manganese, and contain an enzyme called bromelain, which aids in digestion and can tenderize meat.\nSymbolism: The pineapple is a traditional symbol of hospitality and welcome in many cultures.\nAre you asking about the fruit itself, or its use in a specific context?",  
            "forget_at": "None",  
            "invalid_at": "None",  
            "memory_id": "6c8983badede11f083f184ba59bc53c7",  
            "message_id": 269,  
            "message_type": "raw",  
            "session_id": "bf0a50abeb8111f0917884ba59bc53c7",  
            "source_id": "None",  
            "status": true,  
            "user_id": "",  
            "valid_at": "2026-01-07 16:49:12"  
        },  
        ...other 9 messages here  
    ],  
    "message": true  
}  

```

--------------------------------

### Start and verify the Go deployment

Source: https://ragflow.io/docs/migration

Start the Go stack with the target configuration and the same Compose project name used for its volumes, then verify the running containers. Check container logs for migration errors before using the service.

```bash
docker compose --env-file docker/.env -f docker/docker-compose.yml up -d  
docker compose --env-file docker/.env -f docker/docker-compose.yml ps  

```

--------------------------------

### CREATE USER Example

Source: https://ragflow.io/docs/admin_cli

Example invocation of the CREATE USER command. Returns SUCCESS on success.

```sql
RAGFlow(admin)> CREATE USER 'alice@example.com' 'Alice@123456';
SUCCESS
```

--------------------------------

### Update chunk content example

Source: https://ragflow.io/docs/python_api_reference

Example of updating a chunk's content. It initializes the RAGFlow client, retrieves a dataset by ID, gets a document by ID, adds a chunk, and then updates the chunk's content with a new string.

```python
from ragflow_sdk import RAGFlow  
  
rag_object = RAGFlow(api_key="<YOUR_API_KEY>", base_url="http://<YOUR_BASE_URL>:9380")  
dataset = rag_object.list_datasets(id="123")  
dataset = dataset[0]  
doc = dataset.list_documents(id="wdfxb5t547d")  
doc = doc[0]  
chunk = doc.add_chunk(content="xxxxxxx")  
chunk.update({"content":"sdfx..."})  
```

--------------------------------

### Start Go deployment and check services after switching document engine

Source: https://ragflow.io/docs/faq

Starts the Go Compose deployment and lists its services to verify the Infinity document engine is running. Run after setting DOC_ENGINE=infinity.

```bash
docker compose --env-file docker/.env -f docker/docker-compose.yml up -d  
docker compose --env-file docker/.env -f docker/docker-compose.yml ps  
```

--------------------------------

### HELP syntax

Source: https://ragflow.io/docs/admin_cli

Shows CLI help. Multiple aliases are accepted.

```sql
\?  
\h  
\help
```

--------------------------------

### Get Memory Config API Request

Source: https://ragflow.io/docs/http_api_reference

Retrieves the configuration of a specific memory by its memory_id. The example uses memory_id 6c8983badede11f083f184ba59bc53c7. Requires an Authorization header.

```bash
curl --location 'http://{address}/api/v1/memories/6c8983badede11f083f184ba59bc53c7/config' \  
--header 'Authorization: Bearer <YOUR_API_KEY>'  
```

--------------------------------

### Search messages with curl

Source: https://ragflow.io/docs/http_api_reference

Searches messages in a memory using the GET /api/v1/messages/search endpoint. Requires the Authorization header with a valid API key. The query is URL-encoded and the example uses similarity_threshold=0.2, keywords_similarity_weight=0.7, and top_n=10.

```bash
curl --location 'http://{address}/api/v1/messages/search?query=%22who%20are%20you%3F%22&memory_id=6c8983badede11f083f184ba59bc53c7&similarity_threshold=0.2&keywords_similarity_weight=0.7&top_n=10' \  
--header 'Authorization: Bearer <YOUR_API_KEY>'  
```

--------------------------------

### SHOW API SERVER Example

Source: https://ragflow.io/docs/admin_cli

Example invocation of the SHOW API SERVER command. If the name does not exist, returns 'api_server=N/A'.

```sql
RAGFlow(admin)> SHOW API SERVER 'default';
```

--------------------------------

### Install specific RAGFlow CLI version on Linux/macOS

Source: https://ragflow.io/docs/admin_cli

Installs a specific version of the CLI by passing VERSION to the installer. Useful for production or reproducible installations to avoid later releases changing the installed version.

```bash
curl -fsSL https://raw.githubusercontent.com/infiniflow/ragflow/main/tools/scripts/install.sh \
  | VERSION=v1.0.0-rc1 sh  

```

--------------------------------

### Install specific RAGFlow CLI version on Windows PowerShell

Source: https://ragflow.io/docs/admin_cli

Downloads the installer script and passes -Version to install a specific version. The installer installs ragflow-cli.exe under %LOCALAPPDATA%\Programs\RAGFlow and adds that directory to the user PATH.

```powershell
irm https://raw.githubusercontent.com/infiniflow/ragflow/main/tools/scripts/install.ps1 -OutFile install.ps1  
./install.ps1 -Version v1.0.0-rc1  

```

--------------------------------

### Initialize environment variables

Source: https://ragflow.io/docs/sandbox_quickstart

Initializes environment variables by copying the example environment file to .env.

```bash
cp .env.example .env  
```

--------------------------------

### Verify RAGFlow CLI installation

Source: https://ragflow.io/docs/admin_cli

Checks the installed CLI version to confirm the installation succeeded.

```bash
ragflow-cli --version  

```

--------------------------------

### SHOW SERVICE Example

Source: https://ragflow.io/docs/admin_cli

Example of showing the status of a specific service, here 'mysql'.

```text
RAGFlow(admin)> SHOW SERVICE 'mysql';  
```
Tokens:
Raw