> ## Documentation Index > Fetch the complete documentation index at: https://docs.ironbee.ai/llms.txt > Use this file to discover all available pages before exploring further. > ## Agent Instructions > IronBee (ironbee.ai) is an AI QA engineer: it verifies code changes against the running app and keeps the evidence. It is not related to the IronBee open-source web application firewall. > The console is at https://console.ironbee.ai. > Install the CLI with `npm install -g @ironbee-ai/cli`, sign in with `ironbee login`, and set up a project with `ironbee install`. > In GitHub workflows, pin the action to `ironbee-ai/ironbee-action@v1.1.0` and pass the `IRONBEE_API_KEY` secret as the `ironbee_api_key` input. > Console scenarios (https://docs.ironbee.ai/console/scenarios) and CLI saved scenarios (https://docs.ironbee.ai/cli/guides/scenarios) are separate features. > For plans and prices, link to https://ironbee.ai/pricing; the website is the source of truth for them. # Inputs, references and imports > Pass values into a scenario, use variables and secrets in its actions, and reuse one scenario in another. Most [scenarios](/console/scenarios) need values from outside their own rows: a secret to sign in with, a URL that differs by environment, a price read on one page and checked on the next, or a sign-in flow that another scenario already has. *** ## Inputs An input is a value a run passes in, such as a product name or a search term. Add inputs in **Scenario settings** with **Add input**: | Field | What it's for | | - | - | | **Name** | How actions read it, with `@`. Letters, digits, underscore and dash | | **Description** | What the value is | | **Default value** | Used when a run doesn't pass one. Leave it empty and the input is required | An input without a default is required: **Run scenario** asks for it, and every scenario that imports this one must pass it. Renaming an input updates every row that reads it. When you [run a scenario](/console/scenario-runs#start-a-run), you type each input as plain text. *** ## Saved values An **Extract** reads a value from the app and saves it under the names in its **Save as** field. The actions after it can read the value with `@`. For example, an Extract saves `firstPrice` from the product list, and a later Assert checks that the cart total equals it. A value is available only to the rows after the one that writes it. If a row reads a value written later, the editor warns you, and a run that reaches that row ends in a run error. *** ## Outputs Outputs are the values a scenario gives back to the scenarios that import it. Add them in **Scenario settings** with **Add output**. An output is a value from an Extract in **Main**, or an output of an import in **Main**. *** ## References In an instruction, type `@` (or `{{`) at the start of a word. A picker lists everything the action can use, grouped as: * **Inputs** * **Saved values** * **Outputs of imported scenarios** * **Variables** * **Secrets**, with their fields * **Environment properties** Each entry says where it comes from, such as the row that writes a value, or the environments that define a variable. Keep typing to filter, then pick one with the arrow keys and Enter. The pick shows as one token in the instruction. Hover a token to see where its value comes from. Reference picker listing an input with its default, the outputs of imported scenarios with the section they come from, and variables with the environments that define them Reference picker listing an input with its default, the outputs of imported scenarios with the section they come from, and variables with the environments that define them How a run gets the value: * A variable's or property's value is written into the action as text. * A secret is passed as a reference. Its value is never written into the agent's instructions. See [Secrets](/console/secrets#how-a-run-uses-a-secret). * An **HTTP headers** secret is applied to every request automatically, so it can't be used in an action. The picker lists it, turned off. Variables, secrets and properties come from the environment you run against. The editor warns you when a scenario reads one that no environment defines, or that some environments don't, and the [run dialog](/console/scenario-runs#start-a-run) checks the environment you pick. *** ## Imports An **Import** row runs another scenario of the same project in this one. Use it to write a flow once, such as signing in or resetting the app, and reuse it in the project's other scenarios. 1. Add an **Import** row. The **Import a scenario** picker lists the project's scenarios with the inputs each one needs and the outputs it gives. 2. Pick a scenario. The row's dialog opens. 3. Set **Name in this scenario**. You read its outputs with it, for example `@customer.pageTitle`. 4. Fill its **Inputs**, each with a value or one reference: an input, a saved value, an output or a secret. Leave an optional input empty to use its default. A required one must be filled. 5. Click **Add**. Import dialog for Sign in as customer: its account input set to the customer secret, the pageTitle and brandName outputs it gives back, its name in this scenario, and How it works Import dialog for Sign in as customer: its account input set to the customer secret, the pageTitle and brandName outputs it gives back, its name in this scenario, and How it works How an import runs: * Only its **Main** runs. Its own **Setup** and **Teardown** don't. * Values go in only through its inputs, and come back only through its outputs. * The imported scenario is in the same project and of the same type. The run shares its secrets and variables. * When you edit it, every scenario that imports it runs the new version from its next run. On the canvas, click **Show contents** on an Import row to see what it runs. Its actions count toward the [limits](/console/scenario-editor#limits) of the scenario that imports it. The picker lists the scenarios you can't import under **Can't be imported here**, with the reason: for example, it's the scenario you're editing, it runs another verification type, it would import this one back, or it would bring more actions than the scenario has room for. ### Reusable parts A scenario that another scenario imports is a reusable part. Its **Used By** count on the [scenarios list](/console/scenarios#a-projects-scenarios), and **Used by** on its page, show which scenarios import it. * You can't delete a scenario while another one imports it. * A save that would break a scenario that imports it is refused, for example removing an output it reads. The editor names the scenario it would break. *** ## What's next? Pick an environment, pass inputs and read the run. Credentials a run can use without seeing them. This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.