Can I use Mistralai as a CLI?

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

Can I expose Mistralai through MCP?

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

Where are Mistralai credentials stored?

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

Can I deny write access for Mistralai MCP tools?

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

other

Mistralai CLI for AI Agents

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

7 functions 3 read 4 write API key auth

Submit Mistralai feedback View package source

Mistralai CLI Setup

Mistralai can be configured headlessly with `kosmokrator integrations:configure mistralai`.

# 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 mistralai --set api_key="$MISTRALAI_API_KEY" --enable --read allow --write ask --json
kosmokrator integrations:doctor mistralai --json
kosmokrator integrations:status --json

Credentials

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

Key	Env var	Type	Required	Label
`api_key`	`MISTRALAI_API_KEY`	Secret `secret`	yes	API Key
`url`	`MISTRALAI_URL`	URL `url`	no	API Base URL

Call Mistralai Headlessly

Use the generic call form when another coding CLI or script needs a stable universal interface.

kosmo integrations:call mistralai.mistralai_chat '{
  "model": "example_model",
  "messages": "example_messages",
  "temperature": 1,
  "max_tokens": 1
}' --json

Use the provider shortcut form for shorter human-facing commands.

kosmo integrations:mistralai mistralai_chat '{
  "model": "example_model",
  "messages": "example_messages",
  "temperature": 1,
  "max_tokens": 1
}' --json

Agent Discovery Commands

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

kosmo integrations:docs mistralai --json
kosmo integrations:docs mistralai.mistralai_chat --json
kosmo integrations:schema mistralai.mistralai_chat --json
kosmo integrations:search "Mistralai" --json
kosmo integrations:list --json

All CLI Functions

Every function below can be called headlessly. The generic form is stable across all integrations; the provider shortcut is shorter but specific to Mistralai.

`mistralai.mistralai_chat`

Write write

Generate a chat completion using a MistralAI model. Send a list of messages (with roles "system", "user", or "assistant") and receive a model-generated response. Use temperature to control creativity (0 = deterministic, 1 = creative).

Parameters: model, messages, temperature, max_tokens

Generic CLI call

kosmo integrations:call mistralai.mistralai_chat '{"model":"example_model","messages":"example_messages","temperature":1,"max_tokens":1}' --json

Provider shortcut

kosmo integrations:mistralai mistralai_chat '{"model":"example_model","messages":"example_messages","temperature":1,"max_tokens":1}' --json

`mistralai.mistralai_create_embedding`

Write write

Generate text embeddings using a MistralAI embedding model. Converts text into numerical vectors for semantic search, similarity comparison, or clustering. Supports single strings or arrays of strings.

Parameters: model, input

Generic CLI call

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

Provider shortcut

kosmo integrations:mistralai mistralai_create_embedding '{"model":"example_model","input":"example_input"}' --json

`mistralai.mistralai_list_models`

Read read

List all models available in your MistralAI account. Returns model IDs, creation timestamps, and capabilities. Use this to discover which models you can use for chat completions or embeddings.

Parameters: none

Generic CLI call

kosmo integrations:call mistralai.mistralai_list_models '{}' --json

Provider shortcut

kosmo integrations:mistralai mistralai_list_models '{}' --json

`mistralai.mistralai_finetune`

Write write

Create a fine-tuning job on MistralAI. Upload training data and select a base model to create a custom fine-tuned model. The job runs asynchronously — check the returned job ID for status updates.

Parameters: model, training_files, hyperparameters, suffix

Generic CLI call

kosmo integrations:call mistralai.mistralai_finetune '{"model":"example_model","training_files":"example_training_files","hyperparameters":"example_hyperparameters","suffix":"example_suffix"}' --json

Provider shortcut

kosmo integrations:mistralai mistralai_finetune '{"model":"example_model","training_files":"example_training_files","hyperparameters":"example_hyperparameters","suffix":"example_suffix"}' --json

`mistralai.mistralai_list_agents`

Read read

List all MistralAI agents in your account. Returns agent IDs, names, models, and descriptions. Agents are AI assistants with custom instructions and tools.

Parameters: none

Generic CLI call

kosmo integrations:call mistralai.mistralai_list_agents '{}' --json

Provider shortcut

kosmo integrations:mistralai mistralai_list_agents '{}' --json

`mistralai.mistralai_create_agent`

Write write

Create a new MistralAI agent. Specify a name, model, and instructions to define how the agent should behave. Agents are persistent AI assistants that can be used for conversations with custom personalities and capabilities.

Parameters: name, model, instructions, description

Generic CLI call

kosmo integrations:call mistralai.mistralai_create_agent '{"name":"example_name","model":"example_model","instructions":"example_instructions","description":"example_description"}' --json

Provider shortcut

kosmo integrations:mistralai mistralai_create_agent '{"name":"example_name","model":"example_model","instructions":"example_instructions","description":"example_description"}' --json

`mistralai.mistralai_get_current_user`

Read read

Get information about the currently authenticated MistralAI user. Returns account details, subscription tier, and usage information.

Parameters: none

Generic CLI call

kosmo integrations:call mistralai.mistralai_get_current_user '{}' --json

Provider shortcut

kosmo integrations:mistralai mistralai_get_current_user '{}' --json

Function Schemas

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

`mistralai.mistralai_chat`

Operation: Write write
Schema command: kosmo integrations:schema mistralai.mistralai_chat --json

Parameter	Type	Required	Description
`model`	string	yes	The model to use (e.g., "mistral-large-latest", "mistral-small-latest", "open-mistral-nemo").
`messages`	array	yes	Array of message objects with "role" (system, user, assistant) and "content" fields. Example: [{"role": "user", "content": "Hello"}]
`temperature`	number	no	Sampling temperature between 0.0 and 1.0. Lower values are more deterministic, higher values more creative. Default: 0.7.
`max_tokens`	integer	no	Maximum number of tokens to generate in the response.

`mistralai.mistralai_create_embedding`

Operation: Write write
Schema command: kosmo integrations:schema mistralai.mistralai_create_embedding --json

Parameter	Type	Required	Description
`model`	string	yes	The embedding model to use (e.g., "mistral-embed").
`input`	string	yes	The text to embed. Can be a single string or JSON array of strings for batch embedding.

`mistralai.mistralai_list_models`

List all models available in your MistralAI account. Returns model IDs, creation timestamps, and capabilities. Use this to discover which models you can use for chat completions or embeddings.

Operation: Read read
Schema command: kosmo integrations:schema mistralai.mistralai_list_models --json

Parameter	Type	Required	Description
No parameters.

`mistralai.mistralai_finetune`

Create a fine-tuning job on MistralAI. Upload training data and select a base model to create a custom fine-tuned model. The job runs asynchronously — check the returned job ID for status updates.

Operation: Write write
Schema command: kosmo integrations:schema mistralai.mistralai_finetune --json

Parameter	Type	Required	Description
`model`	string	yes	The base model to fine-tune (e.g., "open-mistral-7b", "open-mixtral-8x7b").
`training_files`	array	yes	Array of training file IDs (previously uploaded to MistralAI).
`hyperparameters`	string	no	JSON-encoded hyperparameters object (e.g., {"n_epochs": 3, "learning_rate": 0.0001}).
`suffix`	string	no	Suffix for the fine-tuned model name.

`mistralai.mistralai_list_agents`

List all MistralAI agents in your account. Returns agent IDs, names, models, and descriptions. Agents are AI assistants with custom instructions and tools.

Operation: Read read
Schema command: kosmo integrations:schema mistralai.mistralai_list_agents --json

Parameter	Type	Required	Description
No parameters.

`mistralai.mistralai_create_agent`

Operation: Write write
Schema command: kosmo integrations:schema mistralai.mistralai_create_agent --json

Parameter	Type	Required	Description
`name`	string	yes	The agent name (e.g., "Support Bot", "Code Assistant").
`model`	string	yes	The model to use for the agent (e.g., "mistral-large-latest", "mistral-small-latest").
`instructions`	string	yes	System instructions that define the agent's behavior, personality, and constraints.
`description`	string	no	A short description of what the agent does.

`mistralai.mistralai_get_current_user`

Get information about the currently authenticated MistralAI user. Returns account details, subscription tier, and usage information.

Operation: Read read
Schema command: kosmo integrations:schema mistralai.mistralai_get_current_user --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.