OpenAI#

Access OpenAI's platform API with API key authentication. Manage files, assistants, vector stores, batches, fine-tuning jobs, and model resources. Generate chat completions, images, audio, and video.

This skill uses ClawLink for hosted connection flows and credentials so you do not need to configure OpenAI API access yourself.

Setup in 3 Steps#

Step 1: InstallStep 2: Pair AccountStep 3: Connect OpenAI_App-specific connection GIF coming soon_Run the install command in OpenClawSign in and approve the deviceOpen the dashboard and connect OpenAI

How It Works#

text

┌─────────────────┐     ┌──────────────┐     ┌──────────────────┐
│   OpenClaw      │────▶│   ClawLink   │────▶│   OpenAI API     │
│   (User Chat)   │     │   (API Key)  │     │                  │
└─────────────────┘     └──────────────┘     └──────────────────┘
          │                       │                       │
          │  1. Install Plugin   │                       │
          │  2. Pair Device      │                       │
          │3. Connect OpenAI   │                       │
          │                      │  4. Secure Proxy      │
          │                      │  5. API Requests      │
          │                      │                       │
          ▼                      ▼                       ▼
    ┌──────────┐           ┌──────────┐           ┌──────────┐
    │  SKILL   │           │ Dashboard│           │  OpenAI  │
    │  File    │           │ Auth     │           │ Platform │
    └──────────┘           └──────────┘           └──────────┘

Install#

Install the verified ClawLink plugin from ClawHub. After the user confirms, complete the full setup so ClawLink tools are visible in the next OpenClaw chat.

bash

openclaw plugins install clawhub:clawlink-plugin
openclaw config set tools.alsoAllow '["clawlink-plugin"]' --strict-json
openclaw gateway restart

Then tell the user: "OpenClaw has been restarted. Send /new as a standalone message to start a fresh chat, then ask for OpenAI again."

Quick Start#

bash

# List assistants
clawlink_call_tool --tool "openai_list_assistants" --params '{"limit": 10}'

# Create a chat completion
clawlink_call_tool --tool "openai_create_chat_completion" --params '{"model": "gpt-4o", "messages": [{"role": "user", "content": "Hello"}]}'

# List files
clawlink_call_tool --tool "openai_list_files" --params '{"purpose": "batch"}'

Authentication#

All OpenAI tool calls are authenticated automatically by ClawLink using the user's OpenAI API key.

No API key is required in chat. ClawLink stores the API key securely and injects it into every OpenAI API request on the user's behalf.

Getting Connected#

1. Install the ClawLink plugin (see Install above).
2. Pair the plugin with clawlink_begin_pairing if it is not configured yet.
3. Open https://claw-link.dev/dashboard?add=openai and connect OpenAI.
4. Call clawlink_list_integrations to verify the connection is active.

Connection Management#

List Connections#

bash

clawlink_list_integrations

Response: Returns all connected integrations. Look for openai in the list.

Verify Connection#

bash

clawlink_list_tools --integration openai

Response: Returns the live tool catalog for OpenAI.

Reconnect#

If OpenAI tools are missing or the connection shows an error:

1. Direct the user to https://claw-link.dev/dashboard?add=openai
2. After they confirm, call clawlink_list_integrations to verify
3. Then call clawlink_list_tools --integration openai

Security& Permissions#

• Access is scoped to resources within the connected OpenAI organization and account.
• All write operations require explicit user confirmation. Before executing any create, update, or delete call, confirm the target resource and intended effect with the user.
• Destructive actions (delete file, delete assistant, cancel batch) must be confirmed.
• Fine-tuning, batch jobs, and video generation are asynchronous — poll for completion.

Tool Reference#

Assistants & Threads#

ToolDescriptionModeopenai_list_assistantsList assistants to discover by name or metadataReadopenai_create_assistantCreate a new assistant with model and toolsWriteopenai_modify_assistantUpdate an existing assistantWriteopenai_delete_assistantDelete an assistant permanentlyWriteopenai_create_threadCreate a new conversation threadWriteopenai_create_messageAdd a message to a threadWriteopenai_create_runExecute an assistant on a threadWriteopenai_list_messagesList messages in a threadReadopenai_list_runsList runs for a threadReadopenai_cancel_runCancel an in-progress runWriteopenai_list_run_stepsList steps in a runReadopenai_get_run_stepGet details of a specific run stepRead

Files & Vector Stores#

ToolDescriptionModeopenai_list_filesList uploaded filesReadopenai_upload_fileUpload a file to OpenAIWriteopenai_delete_fileDelete an uploaded fileWriteopenai_download_fileDownload file contentReadopenai_create_vector_storeCreate a vector storeWriteopenai_list_vector_storesList vector storesReadopenai_get_vector_storeGet vector store detailsReadopenai_delete_vector_storeDelete a vector storeWriteopenai_create_vector_store_fileAdd a file to a vector storeWriteopenai_list_vector_store_filesList files in a vector storeReadopenai_delete_vector_store_fileRemove a file from a vector storeWrite

Batches & Fine-Tuning#

ToolDescriptionModeopenai_list_batchesList batch jobsReadopenai_create_batchCreate a batch from a JSONL fileWriteopenai_cancel_batchCancel an in-progress batchWriteopenai_list_fine_tunesList fine-tuning jobsReadopenai_create_fine_tuning_jobStart a fine-tuning jobWriteopenai_list_fine_tuning_eventsGet events for a fine-tuning jobReadopenai_list_fine_tuning_job_checkpointsList checkpoints from a fine-tuning jobRead

Generations & Completions#

ToolDescriptionModeopenai_create_chat_completionCreate a chat completionReadopenai_create_completionCreate a text completion (legacy)Readopenai_create_responseCreate a response via Responses APIWriteopenai_create_embeddingsGenerate text embeddingsReadopenai_create_imageGenerate an imageWriteopenai_create_image_editEdit an existing imageWriteopenai_create_image_variationCreate an image variationWriteopenai_create_speechGenerate text-to-speech audioReadopenai_create_audio_transcriptionTranscribe audio to textReadopenai_create_audio_translationTranslate audio to EnglishReadopenai_create_videoGenerate a videoWriteopenai_list_videosList video generation jobsReadopenai_get_videoGet video generation statusReadopenai_download_videoDownload video contentReadopenai_create_moderationClassify content for policy violationsRead

Evaluations#

ToolDescriptionModeopenai_list_evalsList evaluationsReadopenai_get_evalGet evaluation detailsReadopenai_create_evalCreate an evaluationWriteopenai_delete_evalDelete an evaluationWriteopenai_create_eval_runStart an evaluation runWriteopenai_get_eval_runGet evaluation run statusReadopenai_list_eval_runsList evaluation runsReadopenai_cancel_eval_runCancel an evaluation runWrite

Other Resources#

ToolDescriptionModeopenai_list_modelsList available modelsReadopenai_list_enginesList available enginesReadopenai_get_input_token_countsCalculate token counts for a requestReadopenai_create_skillCreate a skill from filesWriteopenai_list_skillsList skillsReadopenai_delete_skillDelete a skillWriteopenai_create_containerCreate an isolated execution containerWriteopenai_list_containersList containersReadopenai_delete_containerDelete a containerWrite

Code Examples#

Create a chat completion#

bash

clawlink_call_tool --tool "openai_create_chat_completion" \
  --params '{
    "model": "gpt-4o",
    "messages": [
      {"role": "system", "content": "You are a helpful assistant."},
      {"role": "user", "content": "Explain quantum computing in simple terms."}
    ],
    "max_tokens": 500
  }'

Create an assistant#

bash

clawlink_call_tool --tool "openai_create_assistant" \
  --params '{
    "name": "Code Reviewer",
    "model": "gpt-4o",
    "instructions": "You review code and suggest improvements.",
    "tools": [{"type": "code_interpreter"}]
  }'

Upload a file and create a vector store#

bash

clawlink_call_tool --tool "openai_upload_file" \
  --params '{
    "file": "@./knowledge.pdf",
    "purpose": "assistants"
  }'

clawlink_call_tool --tool "openai_create_vector_store_file" \
  --params '{
    "vector_store_id": "VS_ID",
    "file_id": "FILE_ID"
  }'

Create a batch job#

bash

clawlink_call_tool --tool "openai_create_batch" \
  --params '{
    "input_file_id": "FILE_ID",
    "endpoint": "chat/completions",
    "model": "gpt-4o-mini",
    "completion_window": "24h"
  }'

Generate an image#

bash

clawlink_call_tool --tool "openai_create_image" \
  --params '{
    "model": "dall-e-3",
    "prompt": "A minimalist diagram showing the OpenClaw to ClawLink to Provider flow",
    "size": "1024x1024",
    "n": 1
  }'

Discovery Workflow#

1. Call clawlink_list_integrations to confirm OpenAI is connected.
2. Call clawlink_list_tools --integration openai to see the live catalog.
3. Treat the returned list as the source of truth. Do not guess or assume what tools exist.
4. If the user describes a capability but the exact tool is unclear, call clawlink_search_tools with a short query and integration openai.
5. If no OpenAI tools appear, direct the user to https://claw-link.dev/dashboard?add=openai.

Execution Workflow#

text

┌─────────────────────────────────────────────────────────────┐
│  READ OPERATIONS (Safe)                                     │
│  list → get → search → describe → call                      │
│                                                             │
│  Example: List models → Create completion → Return result   │
└─────────────────────────────────────────────────────────────┘
                              │
                              ▼
┌─────────────────────────────────────────────────────────────┐
│  WRITE OPERATIONS (Require Confirmation)                     │
│  list → get → describe → preview → confirm → call           │
│                                                             │
│  Example: Describe tool → Preview changes → User approves   │
│           → Execute update                                  │
└─────────────────────────────────────────────────────────────┘

1. For unfamiliar tools, ambiguous requests, or any write action, call clawlink_describe_tool first.
2. Use the returned guidance, schema, whenToUse, askBefore, safeDefaults, examples, and followups to shape the call.
3. Prefer read, list, search, and get operations before writes when that reduces ambiguity.
4. For writes or anything marked as requiring confirmation, call clawlink_preview_tool first.
5. Execute with clawlink_call_tool. Pass confirmation only after the preview matches the user's intent.
6. If the tool call fails, report the real error. Do not invent results or restate the failure as a missing capability unless the live catalog supports that conclusion.

Notes#

• Use model IDs exactly as returned by openai_list_models — misspelled or unsupported IDs cause validation errors.
• Asynchronous operations (batch, fine-tune, video) return a job ID — poll get endpoints to check completion.
• Thread IDs must be persisted after creation and passed to all subsequent calls.
• Files uploaded for batch processing must have purpose: "batch"; files for assistants must have purpose: "assistants".
• Video generation is asynchronous — poll openai_get_video until status is completed.

Error Handling#

Status / ErrorMeaningTool not foundThe tool name does not exist in the current catalog. Verify with clawlink_list_tools --integration openai.Missing connectionOpenAI is not connected. Direct the user to https://claw-link.dev/dashboard?add=openai.invalid_request_errorInvalid parameter or model not available. Verify model ID with openai_list_models.RateLimitErrorToo many requests. Wait and retry with exponential backoff.authentication_errorInvalid API key. Reconnect OpenAI at the dashboard.Write rejectedUser did not confirm a write action. Always confirm before executing writes.

Troubleshooting: Tools Not Visible#

1. Check that the ClawLink plugin is installed: bash

   openclaw plugins list
   
   

2. If the plugin is installed but tools are missing, tell the user to send /new as a standalone message to reload the catalog.

3. If a fresh chat does not help, run: bash

   openclaw config set tools.alsoAllow '["clawlink-plugin"]' --strict-json
   openclaw gateway restart
   
   

4. After restart, tell the user to send /new again and retry.

Troubleshooting: Invalid Tool Call#

1. Ensure the integration slug is exactly openai.
2. Use clawlink_describe_tool to verify parameter names and types before calling.
3. For write operations, always call clawlink_preview_tool first.

Resources#

• OpenAI API Reference
• Assistants API Overview
• Fine-tuning Documentation
• Batch API
• ClawLink: https://claw-link.dev/?utm_source=clawhub&utm_medium=referral&utm_content=openai-ai
• ClawLink Docs: https://docs.claw-link.dev/openclaw
• ClawLink Verification: https://claw-link.dev/verify

Related Skills#

• OpenAI — For this skill's native documentation

---

Powered by ClawLink — an integration hub for OpenClaw