- Node.js >= 8.0
- Running on Linux? Depending on your distribution, you will need to run the following command:
Debian/Ubuntu:sudo apt-get install libsecret-1-dev
Red Hat-based:sudo yum install libsecret-devel
Arch Linux:sudo pacman -S libsecret
Eventually, the plan is to have self-contained packages for *nix systems and an installer for Windows with no need for manually installing prerequisites.
- Install the CLI globally:
npm install -g twilio-cli
- Now you can run the CLI from anywhere using the
twilio
command.
- Check if there's a new version:
npm outdated -g twilio-cli
- Update the CLI globally:
npm update -g twilio-cli
Plugins for the CLI can be installed using the twilio plugins
command. Until we are publishing the plugins to npm, they will need to be installed by first cloning the plugin repository locally.
-
Clone the plugin repository. Example debugger plugin.
-
"Install" the plugin referencing your plugin's local folder like so:
twilio plugins:link plugin-debugger cd plugin-debugger npm install
-
Now, you can run your plugin command from the cli:
twilio debugger:logs:list --help
Want to write your own plugin? See this document.
twilio project:add
which is an alias for:
twilio project:add -p default
This is for caching your credentials for your existing Twilio account (aka Project) locally. Note, while you are prompted for your Account SID and Auth Token, these are not saved. An API Key is created (look for "twilio-cli for [username] on [hostname]" in the console) and stored in your system's keychain.
You can also use credentials stored in environment variables:
TWILIO_ACCOUNT_SID
= your Account SID from your consoleTWILIO_API_KEY
= an API Key created in your consoleTWILIO_API_SECRET
= the secret for the API Key
TWILIO_ACCOUNT_SID
= your Account SID from your consoleTWILIO_AUTH_TOKEN
= your Auth Token from your console
NOTE: Option 2 should only be used in cases where you are unable to make use of option 1 (which are uncommon).
The CLI will attempt to load credentials in the following order of priority:
- From the project specified with the
-p
parameter. - From the default project, if it exists.
- From environment variables.
twilio
Lists all available commands.
twilio incoming-phone-number:list
Lists all your phone numbers.
Add --help
to any command to get help (e.g. twilio incoming-phone-number:list --help
)
You can set a webhook on a phone number like so:
twilio incoming-phone-number:update [PN sid or E.164] --sms-url http://url
That sets the primary SMS url. There are also options for setting the voice url, fallback url's, and methods for each. Run twilio incoming-phone-number:update --help
for a full list of options.
When you set a webhook, if you specify a URL that uses the host name of localhost
or 127.0.0.1
, the twilio-cli will automatically create an ngrok tunnel for you and set your webhook to the new ngrok URL. For example:
twilio incoming-phone-number:update [PN sid or E.164] --sms-url http://localhost:5000/handle_sms
All command output is sent to stdout
(whereas logging messages are sent to stderr
).
By default, the output is formatted in human readable form in a columnar format like so:
SID Phone Number Friendly Name
PNxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx +1209242XXXX SIP testing
PNxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx +1646887XXXX Congress hotline
PNxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx +1209337XXXX DAVID'S TWILIO CONTACT
Many list
commands will allow you to specify a --properties
option to specify which columns you'd like to display. For example, to display only the Phone Number and SMS Url columns, you would pass --properties "phoneNumber, smsUrl"
.
Note that currently, the column names must match the JSON property names in the Twilio API.
On any command, you can add -o json
to change the output format to JSON. When you choose JSON, the command will send the entire API response to stdout
as JSON. You can then pipe to tools like jq to parse the JSON.
To change the output format to tab separated values (TSV), add -o tsv
to the command line. This format is useful for loading into spreadsheets or for other machine processing. Like the default, columnar output format, you can use the --properties
option to specify which columns you would like included.
All debug, informational, warning, and error information is sent to stderr
. This is so it can be easily separated from the command output. You can decide what level of logging you'd like by using the -l
option. The valid levels of logging messages are debug
, info
, warn
, error
, and none
.
To store credentials for multiple projects, you can use a shorthand "project id" which is just an easy to remember, short string to identify the project. (If you've used git
before, it's like the name you assign to a remote like "origin".)
When you run twilio project:add
, it stores your credentials under a project called default
. This is the project that will be used for all subsequent commands.
To add a second project after the default project, you can run twilio project:add -p my_other_proj
(using whatever identifier you'd like in place of my_other_proj
). Then, when you run subsequent commands, just include the -p my_other_proj
in the command (e.g. twilio incoming-phone-number:list -p my_other_proj
).
To enable autocomplete of CLI commands in bash or zsh, run:
twilio autocomplete
And follow the instructions.
- Clone this repo.
- From the repo directory, run:
npm install
- Run
./bin/run
from the repo directory to run the CLI.
Please file a GitHub issue in this repository for any feedback you may have.
Moving all related CLI libs to v1.0.0 per npm semver recommendation.
- Moved config from
~/.config/@twilio/cli
(Unix) and%LOCALAPPDATA%\@twilio\cli
(Windows) to~/.twilio
and%USERPROFILE%\.twilio
, respectively.
To copy over your existing configuration, run the following commands (Unix):
mkdir ~/.twilio
cp ~/.config/@twilio/cli/config.json ~/.twilio/config.json
- Added autocomplete support.
- Added support for credentials in environment variables.
MIT