Available Actions

add
Adds two numbers
add
Adds two numbers
copy_file
Copies a file from source to destination
create
Create a new task
decrement_action
Decrements a value by 1
delete
Delete a value from the state at a given path
delete
Delete a task
delete_file
Deletes a file or directory with optional recursive deletion
deregister_action
Deregisters an existing action module
divide
Divides one number by another
divide
Divides one number by another
do_work_action
Simulates the robot performing work tasks
enqueue_action
Enqueues another action based on params
eval
Evaluates a mathematical expression
generate_chat_response
Choose an option and explain why
generate_tool_response
Generate a response using LangChain to coordinate with tools/functions
get
Get a value from the state at a given path
get_boolean_response
Get a true/false answer to a question with explanation
get_chat_response
Get a natural language response from the AI assistant
github_issues_create Github API
Create a new issue on GitHub
github_issues_filter Github API
Filter repository issues on GitHub
github_issues_find Github API
Get a specific issue from GitHub
github_issues_list Github API
List all issues from a GitHub repository
github_issues_update Github API
Update an existing issue on GitHub
idle_action
Simulates the robot doing nothing
increment_action
Increments a value by 1
inspect_action
Inspects a value
instructor_chat_completion
Makes a raw chat completion call using Instructor with structured prompting
kill_process
Terminates a child process
langchain_chat_completion
Chat completion action using Langchain
list_directory
Lists directory contents with optional pattern matching
log_action
Logs a message with a specified level
make_directory
Creates a new directory, optionally creating parent directories
move_action
Moves the robot to a specified location
move_file
Moves a file from source to destination
multiply
Multiplies two numbers
multiply
Multiplies two numbers
noop_action
No operation, returns input unchanged
openai_ex_chat_completion
Chat completion using OpenAI Ex with support for tool calling
openai_ex_embeddings
Generate embeddings using OpenAI Ex with support for OpenRouter
openai_ex_image_generation
Generate images using OpenAI Ex with support for OpenRouter
openai_ex_response_retrieve
Retrieve an asynchronous response using OpenAI Ex
random_sleep_action
Introduces a random sleep within a specified range
read_file
Reads content from a file
recharge
Simulates recharging the robot's battery
register_action
Registers a new action module
report_action
Simulates the robot reporting its status
set
Set a value in the state at a given path
sleep_action
Sleeps for a specified duration
spawn_process
Spawns a child process under the agent's supervisor
square
Squares a number
square
Squares a number
state_manager
Generic actions for state management
subtract
Subtracts one number from another
subtract
Subtracts one number from another
tasks
Actions for managing a list of tasks
today
Returns today's date in specified format
todo_action
A placeholder for a todo item
toggle
Toggle the completion status of a task
update
Update a value in the state at a given path with a new value
update
Update an existing task
weather Weather
Get the weather for a given location via the OpenWeatherMap API
write_file
Writes content to a file, optionally creating parent directories

openai_ex_embeddings

Generate embeddings using OpenAI Ex with support for OpenRouter

Jido.AI.Actions.OpenaiEx.Embeddings
Category
Tags

Parameters

model
Required
{:custom, Jido.AI.Model, :validate_model_opts, []}

The AI model to use (e.g., {:openai, [model: "text-embedding-ada-002"]} or %Jido.AI.Model{})

Validation: type: {:custom, Jido.AI.Model, :validate_model_opts, []}
input
Required
{:or, [:string, {:list, :string}]}

The text to generate embeddings for. Can be a single string or a list of strings.

Validation: type: {:or, [:string, {:list, :string}]}
dimensions
integer

The number of dimensions for the embeddings (only supported by some models)

Validation: type: integer
encoding_format
one of [:float, :base64]

The format to return the embeddings in

Default: :float
Validation: type: one of [:float, :base64], must be one of: :float, :base64