> ## Documentation Index > Fetch the complete documentation index at: https://pipedream.com/docs/llms.txt > Use this file to discover all available pages before exploring further. # Bash Prefer to write quick scripts in Bash? We’ve got you covered. You can run any Bash in a Pipedream step within your workflows. Within a Bash step, you can [share data between steps](/workflows/building-workflows/code/bash/#sharing-data-between-steps) and [access environment variables](/workflows/building-workflows/code/bash/#using-environment-variables). But you can’t connect accounts, return HTTP responses, or take advantage of other features available in the [Node.js](/workflows/building-workflows/code/nodejs/) environment at this time. ## Adding a Bash code step 1. Click the + icon to add a new step 2. Click **Custom Code** 3. In the new step, select the `bash` runtime in language dropdown ## Logging and debugging When it comes to debugging Bash scripts, `echo` is your friend. Your `echo` statements will print their output in the workflow step results: ```php theme={null} MESSAGE='Hello world' # The message will now be available in the "Result > Logs" area in the workflow step echo $MESSAGE ``` ## Available binaries Bash steps come with many common and useful binaries preinstalled and available in `$PATH` for you to use out of the box. These binaries include but aren’t limited to: * `curl` for making HTTP requests * `jq` for manipulating and viewing JSON data * `git` for interacting with remote repositories If you need a package pre-installed in your Bash steps, [just ask us](https://pipedream.com/support). Otherwise, you can use the `/tmp` directory to download and install software from source. ## Sharing data between steps A step can accept data from other steps in the same workflow, or pass data downstream to others. ### Using data from another step In Bash steps, data from the initial workflow trigger and other steps are available in the `$PIPEDREAM_STEPS` environment variable. In this example, we’ll pretend this data is coming into our HTTP trigger via a POST request. ```json theme={null} { "id": 1, "name": "Bulbasaur", "type": "plant" } ``` In our Bash script, we can access this data via the `$PIPEDREAM_STEPS` file. Specifically, this data from the POST request into our workflow is available in the `trigger` object. ```go theme={null} cat $PIPEDREAM_STEPS | jq .trigger.event # Results in { id: 1, name: "Bulbasaur", type: "plant" } ``` The period (`.`) in front the `trigger.event` in the example is not a typo. This is to define the starting point for `jq` to traverse down the JSON in the HTTP response. ### Sending data downstream to other steps To share data for future steps to use downstream, append it to the `$PIPEDREAM_EXPORTS` file. ```php theme={null} # Retrieve the data from an API and store it in a variable DATA=`curl --silent https://pokeapi.co/api/v2/pokemon/charizard` # Write data to $PIPEDREAM_EXPORTS to share with steps downstream EXPORT="key:json=${DATA}" echo $EXPORT >> $PIPEDREAM_EXPORTS ``` Not all data types can be stored in the `$PIPEDREAM_EXPORTS` data shared between workflow steps. You can only export JSON-serializable data from Bash steps. ## Using environment variables You can leverage any [environment variables defined in your Pipedream account](/workflows/environment-variables/) in a bash step. This is useful for keeping your secrets out of code as well as keeping them flexible to swap API keys without having to update each step individually. To access them, just append the `$` in front of the environment variable name. ```php theme={null} echo $POKEDEX_API_KEY ``` Or an even more useful example, using the stored environment variable to make an authenticated API request. ```bash theme={null} curl --silent -X POST -h "Authorization: Bearer $TWITTER_API_KEY" https://api.twitter.com/2/users/@pipedream/mentions ``` ## Making a `GET` request You can use `curl` to perform `GET` requests. ```php theme={null} # Get the current weather in San Francisco WEATHER=`curl --silent https://wttr.in/San\ Francisco\?format=3` echo $WEATHER # Produces: # San Francisco: 🌫 +48°F ``` Use the `--silent` flag with `curl` to suppress extra diagnostic information that `curl` produces when making requests. This enables you to only worry about the body of the response so you can visualize it with tools like `echo` or `jq`. ## Making a `POST` request ```php theme={null} curl --silent -X POST https://postman-echo.com/post -d 'name=Bulbasaur&id=1' # To store the API response in a variable, interpolate the response into a string and store it in variable RESPONSE=`curl --silent -X POST https://postman-echo.com/post -d 'name=Bulbasaur&id=1'` # Now the response is stored as a variable echo $RESPONSE ``` ## Using API key authentication Some APIs require you to authenticate with a secret API key. `curl` has an `-h` flag where you can pass your API key as a token. For example, here’s how to retrieve mentions from the Twitter API: ```bash theme={null} # Define the "Authorization" header to include your Twitter API key curl --silent -X POST -h "Authorization: Bearer $()" https://api.twitter.com/2/users/@pipedream/mentions ``` ## Raising exceptions You may need to stop your step immediately. You can use the normal `exit` function available in Bash to quit the step prematurely. ```php theme={null} echo "Exiting now!" 1>&2 exit 1 ``` Using `exit` to quit a Bash step early *won’t* stop the execution of the rest of the workflow. Exiting a Bash step will only apply that particular step in the workflow. This will exit the step and output the error message to `stderr` which will appear in the results of the step in the workflow. ## File storage If you need to download and store files, you can write them to the `/tmp` directory. ### Writing a file to /tmp Download a file to `/tmp` using `curl` ```php theme={null} # Download the current weather in Cleveland in PNG format curl --silent https://wttr.in/Cleveland.png --output /tmp/weather.png # Output the contents of /tmp to confirm the file is there ls /tmp ``` The `/tmp` directory does not have unlimited storage. Please refer to the [disk limits](/workflows/limits/#disk) for details.