Can I use OpenAI as a CLI?

Yes. KosmoKrator exposes a OpenAI CLI through kosmo integrations:call and the provider shortcut command. Commands can return JSON for scripts, CI, and coding agents.

Can I expose OpenAI through MCP?

Yes. The KosmoKrator MCP gateway can expose selected OpenAI tools to clients such as Claude Code, Cursor, Codex, and other MCP-compatible tools.

Where are OpenAI credentials stored?

OpenAI credentials are configured in KosmoKrator's local integration settings and reused by CLI, Lua, and MCP gateway surfaces.

Can I deny write access for OpenAI MCP tools?

Yes. Start the gateway with --write=deny for read-only clients, or use ask/allow only for trusted workspaces.

data

OpenAI CLI for AI Agents

Use the OpenAI CLI from KosmoKrator to call OpenAI tools headlessly, return JSON, inspect schemas, and automate workflows from coding agents, scripts, and CI.

OpenAI CLI Setup

OpenAI can be configured headlessly with `kosmokrator integrations:configure openai`.

Install, configure, and verify

# Install KosmoKrator first if it is not available on PATH.
curl -fsSL https://raw.githubusercontent.com/OpenCompanyApp/kosmokrator/main/install.sh | bash

# Configure and verify this integration.
kosmokrator integrations:configure openai --set api_key="$OPENAI_API_KEY" --enable --read allow --write ask --json
kosmokrator integrations:doctor openai --json
kosmokrator integrations:status --json

Credentials

Authentication type: API key api_key. Configure credentials once, then reuse the same stored profile from scripts, coding CLIs, Lua, and MCP.

Key	Env var	Type	Required	Label
`api_key`	`OPENAI_API_KEY`	Secret `secret`	yes	API Key

Command Patterns

The generic command is stable across every integration. The provider shortcut is shorter for humans.

Generic CLI call

kosmo integrations:call openai.openai_chat_completion '{"model":"example_model","messages":"example_messages","temperature":1,"max_tokens":1,"top_p":1,"frequency_penalty":1,"presence_penalty":1,"response_format":"example_response_format"}' --json

Provider shortcut

kosmo integrations:openai openai_chat_completion '{"model":"example_model","messages":"example_messages","temperature":1,"max_tokens":1,"top_p":1,"frequency_penalty":1,"presence_penalty":1,"response_format":"example_response_format"}' --json

Discovery

These commands return structured output for coding agents that need to inspect capabilities before choosing a function.

Discovery commands

kosmo integrations:docs openai --json
kosmo integrations:docs openai.openai_chat_completion --json
kosmo integrations:schema openai.openai_chat_completion --json
kosmo integrations:search "OpenAI" --json
kosmo integrations:list --json

Automation Contexts

The same configured command surface works in these environments. The command does not change unless the host wrapper, credentials, or permissions change.

CI Run integration calls from CI jobs with JSON output, explicit credentials, and predictable exit status. Cron Jobs Schedule repeatable integration workflows from cron while keeping credentials in KosmoKrator config. Shell Scripts Call integration functions from shell scripts with stable JSON input and output. Headless Automation Use KosmoKrator as a non-interactive integration runtime for local automations and wrappers. Coding Agents Let coding agents discover schemas and execute integration functions through CLI commands or MCP.

CLI Functions

Every function below can be called headlessly. Commands are highlighted, copyable, and scroll horizontally when payloads are long.

`openai.openai_chat_completion`

Generate a chat completion using OpenAI GPT models. Send a conversation as an array of messages with role and content.

Write write

Parameters: model, messages, temperature, max_tokens, top_p, frequency_penalty, presence_penalty, response_format

Generic call

kosmo integrations:call openai.openai_chat_completion '{"model":"example_model","messages":"example_messages","temperature":1,"max_tokens":1,"top_p":1,"frequency_penalty":1,"presence_penalty":1,"response_format":"example_response_format"}' --json

Shortcut

kosmo integrations:openai openai_chat_completion '{"model":"example_model","messages":"example_messages","temperature":1,"max_tokens":1,"top_p":1,"frequency_penalty":1,"presence_penalty":1,"response_format":"example_response_format"}' --json

`openai.openai_create_embedding`

Generate an embedding vector for text input using OpenAI embedding models.

Write write

Parameters: model, input

Generic call

kosmo integrations:call openai.openai_create_embedding '{"model":"example_model","input":"example_input"}' --json

Shortcut

kosmo integrations:openai openai_create_embedding '{"model":"example_model","input":"example_input"}' --json

`openai.openai_create_image`

Generate an image from a text prompt using DALL·E.

Write write

Parameters: model, prompt, n, size, quality, style, response_format

Generic call

kosmo integrations:call openai.openai_create_image '{"model":"example_model","prompt":"example_prompt","n":1,"size":"example_size","quality":"example_quality","style":"example_style","response_format":"example_response_format"}' --json

Shortcut

kosmo integrations:openai openai_create_image '{"model":"example_model","prompt":"example_prompt","n":1,"size":"example_size","quality":"example_quality","style":"example_style","response_format":"example_response_format"}' --json

`openai.openai_transcribe_audio`

Transcribe audio into text using OpenAI Whisper. Supports MP3, MP4, WAV, M4A, and WEBM audio files.

Write write

Parameters: file_content, filename, model, language, response_format

Generic call

kosmo integrations:call openai.openai_transcribe_audio '{"file_content":"example_file_content","filename":"example_filename","model":"example_model","language":"example_language","response_format":"example_response_format"}' --json

Shortcut

kosmo integrations:openai openai_transcribe_audio '{"file_content":"example_file_content","filename":"example_filename","model":"example_model","language":"example_language","response_format":"example_response_format"}' --json

`openai.openai_text_to_speech`

Generate speech audio from text using OpenAI TTS. Returns base64-encoded audio content.

Write write

Parameters: model, input, voice, speed, response_format

Generic call

kosmo integrations:call openai.openai_text_to_speech '{"model":"example_model","input":"example_input","voice":"example_voice","speed":1,"response_format":"example_response_format"}' --json

Shortcut

kosmo integrations:openai openai_text_to_speech '{"model":"example_model","input":"example_input","voice":"example_voice","speed":1,"response_format":"example_response_format"}' --json

`openai.openai_create_assistant`

Create an OpenAI assistant with a model, name, instructions, and optional tools.

Write write

Parameters: model, name, description, instructions, tools

Generic call

kosmo integrations:call openai.openai_create_assistant '{"model":"example_model","name":"example_name","description":"example_description","instructions":"example_instructions","tools":"example_tools"}' --json

Shortcut

kosmo integrations:openai openai_create_assistant '{"model":"example_model","name":"example_name","description":"example_description","instructions":"example_instructions","tools":"example_tools"}' --json

`openai.openai_list_assistants`

List all OpenAI assistants available to the authenticated user.

Read read

Parameters: limit

Generic call

kosmo integrations:call openai.openai_list_assistants '{"limit":1}' --json

Shortcut

kosmo integrations:openai openai_list_assistants '{"limit":1}' --json

`openai.openai_create_thread`

Create a conversation thread for the OpenAI Assistants API. Optionally include initial messages.

Write write

Parameters: messages

Generic call

kosmo integrations:call openai.openai_create_thread '{"messages":"example_messages"}' --json

Shortcut

kosmo integrations:openai openai_create_thread '{"messages":"example_messages"}' --json

`openai.openai_add_message_to_thread`

Add a message to an existing conversation thread.

Write write

Parameters: thread_id, role, content

Generic call

kosmo integrations:call openai.openai_add_message_to_thread '{"thread_id":"example_thread_id","role":"example_role","content":"example_content"}' --json

Shortcut

kosmo integrations:openai openai_add_message_to_thread '{"thread_id":"example_thread_id","role":"example_role","content":"example_content"}' --json

`openai.openai_list_thread_messages`

List messages in an OpenAI conversation thread.

Read read

Parameters: thread_id, limit

Generic call

kosmo integrations:call openai.openai_list_thread_messages '{"thread_id":"example_thread_id","limit":1}' --json

Shortcut

kosmo integrations:openai openai_list_thread_messages '{"thread_id":"example_thread_id","limit":1}' --json

`openai.openai_create_run`

Start an assistant run on a thread. Returns the run with its initial status.

Write write

Parameters: thread_id, assistant_id, instructions

Generic call

kosmo integrations:call openai.openai_create_run '{"thread_id":"example_thread_id","assistant_id":"example_assistant_id","instructions":"example_instructions"}' --json

Shortcut

kosmo integrations:openai openai_create_run '{"thread_id":"example_thread_id","assistant_id":"example_assistant_id","instructions":"example_instructions"}' --json

`openai.openai_get_run`

Get the status and details of an assistant run on a thread.

Read read

Parameters: thread_id, run_id

Generic call

kosmo integrations:call openai.openai_get_run '{"thread_id":"example_thread_id","run_id":"example_run_id"}' --json

Shortcut

kosmo integrations:openai openai_get_run '{"thread_id":"example_thread_id","run_id":"example_run_id"}' --json

`openai.openai_upload_file`

Upload a file to OpenAI for use with Assistants, Fine-tuning, or Batch API.

Write write

Parameters: file_content, filename, purpose

Generic call

kosmo integrations:call openai.openai_upload_file '{"file_content":"example_file_content","filename":"example_filename","purpose":"example_purpose"}' --json

Shortcut

kosmo integrations:openai openai_upload_file '{"file_content":"example_file_content","filename":"example_filename","purpose":"example_purpose"}' --json

`openai.openai_list_files`

List files uploaded to OpenAI, optionally filtered by purpose.

Read read

Parameters: purpose, limit

Generic call

kosmo integrations:call openai.openai_list_files '{"purpose":"example_purpose","limit":1}' --json

Shortcut

kosmo integrations:openai openai_list_files '{"purpose":"example_purpose","limit":1}' --json

`openai.openai_list_models`

List all models available from OpenAI, including GPT, DALL·E, Whisper, and embedding models.

Read read

Parameters: none

Generic call

kosmo integrations:call openai.openai_list_models '{}' --json

Shortcut

kosmo integrations:openai openai_list_models '{}' --json

Function Schemas

Use these parameter tables when building CLI payloads without calling integrations:schema first.

openai.openai_chat_completion 8 parameters

Schema command

kosmo integrations:schema openai.openai_chat_completion --json

Parameter	Type	Required	Description
`model`	string	yes	Model ID (e.g., "gpt-4o", "gpt-4o-mini", "gpt-4-turbo").
`messages`	array	yes	Array of message objects, each with "role" (system, user, assistant) and "content".
`temperature`	number	no	Sampling temperature between 0 and 2. Higher values produce more random output.
`max_tokens`	integer	no	Maximum number of tokens to generate in the response.
`top_p`	number	no	Nucleus sampling parameter. Use temperature or top_p, but not both.
`frequency_penalty`	number	no	Penalty for frequent tokens (-2.0 to 2.0).
`presence_penalty`	number	no	Penalty for new tokens (-2.0 to 2.0).
`response_format`	object	no	Response format, e.g. {"type": "json_object"} for JSON output.

openai.openai_create_embedding 2 parameters

Schema command

kosmo integrations:schema openai.openai_create_embedding --json

Parameter	Type	Required	Description
`model`	string	yes	Embedding model ID (e.g., "text-embedding-3-small", "text-embedding-3-large").
`input`	string	yes	Text string or array of strings to embed.

openai.openai_create_image 7 parameters

Schema command

kosmo integrations:schema openai.openai_create_image --json

Parameter	Type	Required	Description
`model`	string	no	Model ID (e.g., "dall-e-3", "dall-e-2"). Default: "dall-e-2".
`prompt`	string	yes	Text description of the desired image.
`n`	integer	no	Number of images to generate (1-10 for DALL·E 2, only 1 for DALL·E 3).
`size`	string	no	Image size: "256x256", "512x512", "1024x1024", "1024x1792", "1792x1024".
`quality`	string	no	Image quality: "standard" or "hd" (DALL·E 3 only).
`style`	string	no	Image style: "vivid" or "natural" (DALL·E 3 only).
`response_format`	string	no	Response format: "url" or "b64_json".

openai.openai_transcribe_audio 5 parameters

Schema command

kosmo integrations:schema openai.openai_transcribe_audio --json

Parameter	Type	Required	Description
`file_content`	string	yes	Base64-encoded audio file content.
`filename`	string	yes	Filename with extension (e.g., "audio.mp3", "recording.wav").
`model`	string	yes	Model to use for transcription (e.g., "whisper-1").
`language`	string	no	ISO 639-1 language code (e.g., "en", "fr", "de").
`response_format`	string	no	Output format: "json", "text", "srt", "verbose_json", "vtt".

openai.openai_text_to_speech 5 parameters

Schema command

kosmo integrations:schema openai.openai_text_to_speech --json

Parameter	Type	Required	Description
`model`	string	yes	TTS model ID (e.g., "tts-1", "tts-1-hd").
`input`	string	yes	Text to convert to speech (max 4096 characters).
`voice`	string	yes	Voice to use: "alloy", "echo", "fable", "onyx", "nova", or "shimmer".
`speed`	number	no	Speed of speech (0.25 to 4.0). Default: 1.0.
`response_format`	string	no	Audio format: "mp3", "opus", "aac", "flac", "wav". Default: "mp3".

openai.openai_create_assistant 5 parameters

Schema command

kosmo integrations:schema openai.openai_create_assistant --json

Parameter	Type	Required	Description
`model`	string	yes	Model ID (e.g., "gpt-4o").
`name`	string	no	Name of the assistant.
`description`	string	no	Description of the assistant.
`instructions`	string	no	System instructions for the assistant.
`tools`	array	no	Array of tool objects the assistant can use (e.g., code_interpreter, file_search).

openai.openai_list_assistants 1 parameters

Schema command

kosmo integrations:schema openai.openai_list_assistants --json

Parameter	Type	Required	Description
`limit`	integer	no	Number of assistants to return (default 20, max 100).

openai.openai_create_thread 1 parameters

Schema command

kosmo integrations:schema openai.openai_create_thread --json

Parameter	Type	Required	Description
`messages`	array	no	Initial messages for the thread. Each message has "role" and "content".

openai.openai_add_message_to_thread 3 parameters

Schema command

kosmo integrations:schema openai.openai_add_message_to_thread --json

Parameter	Type	Required	Description
`thread_id`	string	yes	The ID of the thread to add the message to.
`role`	string	yes	Role of the message sender: "user" or "assistant".
`content`	string	yes	Text content of the message.

openai.openai_list_thread_messages 2 parameters

Schema command

kosmo integrations:schema openai.openai_list_thread_messages --json

Parameter	Type	Required	Description
`thread_id`	string	yes	The ID of the thread to list messages from.
`limit`	integer	no	Number of messages to return (default 20, max 100).

openai.openai_create_run 3 parameters

Schema command

kosmo integrations:schema openai.openai_create_run --json

Parameter	Type	Required	Description
`thread_id`	string	yes	The ID of the thread to run the assistant on.
`assistant_id`	string	yes	The ID of the assistant to use for this run.
`instructions`	string	no	Override the assistant's default instructions for this run.

openai.openai_get_run 2 parameters

Schema command

kosmo integrations:schema openai.openai_get_run --json

Parameter	Type	Required	Description
`thread_id`	string	yes	The ID of the thread.
`run_id`	string	yes	The ID of the run to check.

openai.openai_upload_file 3 parameters

Schema command

kosmo integrations:schema openai.openai_upload_file --json

Parameter	Type	Required	Description
`file_content`	string	yes	Base64-encoded file content.
`filename`	string	yes	Filename with extension (e.g., "data.jsonl", "document.txt").
`purpose`	string	yes	Purpose of the file: "assistants", "assistants_output", "batch", "fine-tune", "vision".

openai.openai_list_files 2 parameters

Schema command

kosmo integrations:schema openai.openai_list_files --json

Parameter	Type	Required	Description
`purpose`	string	no	Filter by purpose: "assistants", "assistants_output", "batch", "fine-tune", "vision".
`limit`	integer	no	Number of files to return (default 20, max 10000).

openai.openai_list_models 0 parameters

Schema command

kosmo integrations:schema openai.openai_list_models --json

Parameter	Type	Required	Description
No parameters.

Permissions

Headless calls still follow the integration read/write permission policy. Configure read/write defaults with integrations:configure. Add --force only for trusted automation that should bypass that policy.