# Send Open WebUI to your agent
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
## Fast path
- Download the package from Yavira.
- Extract it into a folder your agent can access.
- Paste one of the prompts below and point your agent at the extracted folder.
## Suggested prompts
### New install

```text
I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete.
```
### Upgrade existing

```text
I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run.
```
## Machine-readable fields
```json
{
  "schemaVersion": "1.0",
  "item": {
    "slug": "open-webui",
    "name": "Open WebUI",
    "source": "tencent",
    "type": "skill",
    "category": "开发工具",
    "sourceUrl": "https://clawhub.ai/0x7466/open-webui",
    "canonicalUrl": "https://clawhub.ai/0x7466/open-webui",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadUrl": "/downloads/open-webui",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=open-webui",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "packageFormat": "ZIP package",
    "primaryDoc": "SKILL.md",
    "includedAssets": [
      "SKILL.md",
      "scripts/openwebui-cli.py"
    ],
    "downloadMode": "redirect",
    "sourceHealth": {
      "source": "tencent",
      "slug": "open-webui",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-05-06T09:42:34.078Z",
      "expiresAt": "2026-05-13T09:42:34.078Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=open-webui",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=open-webui",
        "contentDisposition": "attachment; filename=\"open-webui-1.0.0.zip\"",
        "redirectLocation": null,
        "bodySnippet": null,
        "slug": "open-webui"
      },
      "scope": "item",
      "summary": "Item download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this item.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/open-webui"
    },
    "validation": {
      "installChecklist": [
        "Use the Yavira download entry.",
        "Review SKILL.md after the package is downloaded.",
        "Confirm the extracted package contains the expected setup assets."
      ],
      "postInstallChecks": [
        "Confirm the extracted package includes the expected docs or setup files.",
        "Validate the skill or prompts are available in your target agent workspace.",
        "Capture any manual follow-up steps the agent could not complete."
      ]
    }
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/open-webui",
    "downloadUrl": "https://openagent3.xyz/downloads/open-webui",
    "agentUrl": "https://openagent3.xyz/skills/open-webui/agent",
    "manifestUrl": "https://openagent3.xyz/skills/open-webui/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/open-webui/agent.md"
  }
}
```
## Documentation

### Open WebUI API Skill

Complete API integration for Open WebUI - a unified interface for LLMs including Ollama, OpenAI, and other providers.

### When to Use

Activate this skill when the user wants to:

List available models from their Open WebUI instance
Send chat completions to models through Open WebUI
Upload files for RAG (Retrieval Augmented Generation)
Manage knowledge collections and add files to them
Use Ollama proxy endpoints (generate, embed, pull models)
Generate images or process audio through Open WebUI
Check Ollama status or manage models (load, unload, delete)
Create or manage pipelines

Do NOT activate for:

Installing or configuring Open WebUI server itself (use system admin skills)
General questions about what Open WebUI is (use general knowledge)
Troubleshooting Open WebUI server issues (use troubleshooting guides)
Local file operations unrelated to Open WebUI API

### Environment Variables (Recommended)

export OPENWEBUI_URL="http://localhost:3000"  # Your Open WebUI instance URL
export OPENWEBUI_TOKEN="your-api-key-here"    # From Settings > Account in Open WebUI

### Authentication

Bearer Token authentication required
Token obtained from Open WebUI: Settings > Account
Alternative: JWT token for advanced use cases

### Activation Triggers

Example requests that SHOULD activate this skill:

"List all models available in my Open WebUI"
"Send a chat completion to llama3.2 via Open WebUI with prompt 'Explain quantum computing'"
"Upload /path/to/document.pdf to Open WebUI knowledge base"
"Create a new knowledge collection called 'Research Papers' in Open WebUI"
"Generate an embedding for 'Open WebUI is great' using the nomic-embed-text model"
"Pull the llama3.2 model through Open WebUI Ollama proxy"
"Get Ollama status from my Open WebUI instance"
"Chat with gpt-4 using my Open WebUI with RAG enabled on collection 'docs'"
"Generate an image using Open WebUI with prompt 'A futuristic city'"
"Delete the old-model from Open WebUI Ollama"

Example requests that should NOT activate this skill:

"How do I install Open WebUI?" (Installation/Admin)
"What is Open WebUI?" (General knowledge)
"Configure the Open WebUI environment variables" (Server config)
"Troubleshoot why Open WebUI won't start" (Server troubleshooting)
"Compare Open WebUI to other UIs" (General comparison)

### 1. Configuration Check

Verify OPENWEBUI_URL and OPENWEBUI_TOKEN are set
Validate URL format (http/https)
Test connection with GET /api/models or /ollama/api/tags

### 2. Operation Execution

Use the CLI tool or direct API calls:

# Using the CLI tool (recommended)
python3 scripts/openwebui-cli.py --help
python3 scripts/openwebui-cli.py models list
python3 scripts/openwebui-cli.py chat --model llama3.2 --message "Hello"

# Using curl (alternative)
curl -H "Authorization: Bearer $OPENWEBUI_TOKEN" \\
  "$OPENWEBUI_URL/api/models"

### 3. Response Handling

HTTP 200: Success - parse and present JSON
HTTP 401: Authentication failed - check token
HTTP 404: Endpoint/model not found
HTTP 422: Validation error - check request parameters

### Chat & Completions

EndpointMethodDescription/api/chat/completionsPOSTOpenAI-compatible chat completions/api/modelsGETList all available models/ollama/api/chatPOSTNative Ollama chat completion/ollama/api/generatePOSTOllama text generation

### Ollama Proxy

EndpointMethodDescription/ollama/api/tagsGETList Ollama models/ollama/api/pullPOSTPull/download a model/ollama/api/deleteDELETEDelete a model/ollama/api/embedPOSTGenerate embeddings/ollama/api/psGETList loaded models

### RAG & Knowledge

EndpointMethodDescription/api/v1/files/POSTUpload file for RAG/api/v1/files/{id}/process/statusGETCheck file processing status/api/v1/knowledge/GET/POSTList/create knowledge collections/api/v1/knowledge/{id}/file/addPOSTAdd file to knowledge base

### Images & Audio

EndpointMethodDescription/api/v1/images/generationsPOSTGenerate images/api/v1/audio/speechPOSTText-to-speech/api/v1/audio/transcriptionsPOSTSpeech-to-text

### Confirmation Required

Always confirm before:

Deleting models (DELETE /ollama/api/delete) - Irreversible
Pulling large models - May take significant time/bandwidth
Deleting knowledge collections - Data loss risk
Uploading sensitive files - Privacy consideration

### Redaction & Security

Never log the full API token - Redact to sk-...XXXX format
Sanitize file paths - Verify files exist before upload
Validate URLs - Ensure HTTPS for external instances
Handle errors gracefully - Don't expose stack traces with tokens

### Workspace Safety

File uploads default to workspace directory
Confirm before accessing files outside workspace
No sudo/root operations required (pure API client)

### List Models

python3 scripts/openwebui-cli.py models list

### Chat Completion

python3 scripts/openwebui-cli.py chat \\
  --model llama3.2 \\
  --message "Explain the benefits of RAG" \\
  --stream

### Upload File for RAG

python3 scripts/openwebui-cli.py files upload \\
  --file /path/to/document.pdf \\
  --process

### Add File to Knowledge Base

python3 scripts/openwebui-cli.py knowledge add-file \\
  --collection-id "research-papers" \\
  --file-id "doc-123-uuid"

### Generate Embeddings (Ollama)

python3 scripts/openwebui-cli.py ollama embed \\
  --model nomic-embed-text \\
  --input "Open WebUI is great for LLM management"

### Pull Model (Confirmation Required)

python3 scripts/openwebui-cli.py ollama pull \\
  --model llama3.2:70b
# Agent must confirm: "This will download ~40GB. Proceed? [y/N]"

### Check Ollama Status

python3 scripts/openwebui-cli.py ollama status

### Error Handling

ErrorCauseSolution401 UnauthorizedInvalid or missing tokenVerify OPENWEBUI_TOKEN404 Not FoundModel/endpoint doesn't existCheck model name spelling422 Validation ErrorInvalid parametersCheck request body format400 Bad RequestFile still processingWait for processing completionConnection refusedWrong URLVerify OPENWEBUI_URL

### File Processing Race Condition

Files uploaded for RAG are processed asynchronously. Before adding to knowledge:

Upload file → get file_id
Poll /api/v1/files/{id}/process/status until status: "completed"
Then add to knowledge collection

### Large Model Downloads

Pulling models (e.g., 70B parameters) can take hours. Always:

Confirm with user before starting
Show progress if possible
Allow cancellation

### Streaming Responses

Chat completions support streaming. Use --stream flag for real-time output or collect full response for non-streaming.

### CLI Tool Reference

The included CLI tool (scripts/openwebui-cli.py) provides:

Automatic authentication from environment variables
Structured JSON output with optional formatting
Built-in help for all commands
Error handling with user-friendly messages
Progress indicators for long operations

Run python3 scripts/openwebui-cli.py --help for full usage.
## Trust
- Source: tencent
- Verification: Indexed source record
- Publisher: 0x7466
- Version: 1.0.0
## Source health
- Status: healthy
- Item download looks usable.
- Yavira can redirect you to the upstream package for this item.
- Health scope: item
- Reason: direct_download_ok
- Checked at: 2026-05-06T09:42:34.078Z
- Expires at: 2026-05-13T09:42:34.078Z
- Recommended action: Download for OpenClaw
## Links
- [Detail page](https://openagent3.xyz/skills/open-webui)
- [Send to Agent page](https://openagent3.xyz/skills/open-webui/agent)
- [JSON manifest](https://openagent3.xyz/skills/open-webui/agent.json)
- [Markdown brief](https://openagent3.xyz/skills/open-webui/agent.md)
- [Download page](https://openagent3.xyz/downloads/open-webui)