What Projects can do
See every capability a Project's agent, its Workflows, and its custom views can use, and how each one reaches it.
Audience: builders
You do not need to write code to use these capabilities. You ask the agent in the Project's conversation, and the agent writes the code that calls them. The names in the tables below help you ask for something precisely, and help you read what the agent did.
Three ways to reach the platform
| Who | How it calls the platform | Where it runs |
|---|---|---|
| The agent | Python code that calls the sdk modules listed on this page | In the Project's sandbox, during a conversation or a Workflow agent step |
| A custom view | Its own runtime calls, such as listDecisions() | In the browser of the person viewing it, with that person's permissions |
| Verified logic | The same sdk modules as the agent, inside reviewed and approved code | From a Workflow step, a conversation, a custom view, or the Project API |
A custom view has its own calls for work items, decisions, files, data reads, and Verified logic. For everything else, including changes to data, a view goes through Verified logic. You ask the agent to write the logic, a reviewer approves it, and a button in the view starts it. This way, every action a view takes beyond its own calls runs reviewed code.
In the tables below, Agent means the agent's code and the code inside Verified logic. View means a custom view's own runtime calls. Where a section has no View calls, a view reaches those capabilities through Verified logic.
Ask a person to decide
A run can stop and wait for a person to choose. The person answers in the conversation, in the work item, or in a custom view.
| Call | What it does | Agent | View |
|---|---|---|---|
sdk.human.ask() | Asks a person a question with options, then ends the script so the run waits for the answer. | Yes | No |
sdk.human.ask_many() | Asks up to four questions at once in a conversation. A Workflow step asks one at a time. | Yes | No |
sdk.meta.work_items.propose_decision() | Proposes an answer to a work item's pending decision for a person to confirm. | Yes | No |
listDecisions() | Lists the Project's decisions, pending by default, newest or oldest first. | No | Yes |
getDecision() | Reads one decision's question, options, pre-filled values, and answer. | No | Yes |
answerDecision() | Answers a decision as the person viewing. | No | Yes |
openDecision() | Opens the work item's decision card over the view, to edit values or add a note. | No | Yes |
sdk.human.ask() ends the script: nothing after the call runs.
The agent writes anything the next run needs to the Project's files before it asks.
A Workflow that must stop for a person every time uses a human.request step instead. See Outcomes.
An answer given in a view is recorded as the person's own. The platform checks it the same way as an answer in the work item, and a decision someone else answered first is not answered again.
Query and change data
These calls work on the data sources attached to the Project, within the access the Project was granted.
| Call | What it does | Agent | View |
|---|---|---|---|
sdk.sql.sources() | Lists the data sources, the Project's access, and the tables and columns it may use. | Yes | No |
sdk.sql.describe_source() | Reads one source's access grant and catalog. | Yes | No |
sdk.sql.list_databases() | Lists the databases an attached connection can reach. | Yes | No |
sdk.sql.read() | Reads data with ANSI SQL over the declared columns. | Yes | No |
sdk.sql.execute() | Runs SQL within the Project's grants. | Yes | No |
sdk.sql.transaction() | Commits 1 to 100 related writes on one source together, or rolls them all back. | Yes | No |
sdk.sql.validate() | Checks SQL against the current grants without sending it to the source. | Yes | No |
sdk.semantic_models | Creates, versions, and binds semantic models, and links Verified logic to them. | Yes | No |
listDataSources() | Lists the data sources attached to the Project. | No | Yes |
openDataSource() | Opens an attached data source in the app. | No | Yes |
readSql() | Reads an attached data source or semantic model with one read-only SQL statement, with the viewer's access. Returns at most 1,000 rows. | No | Yes |
To show query results in a view, call readSql(). To change data from a view, start Verified logic.
Run Workflows and work items
| Call | What it does | Agent | View |
|---|---|---|---|
sdk.meta.workflows.submit_work_item() | Submits one work item to a Workflow. | Yes | No |
sdk.meta.workflows.submit_work_items() | Submits up to 1,000 work items in one call. | Yes | No |
sdk.meta.workflows.poll_work_items() | Reads the progress of up to 1,000 submitted work items. | Yes | No |
sdk.meta.workflows.list_triggers() | Lists a Workflow's triggers. | Yes | No |
sdk.meta.workflows.pause_trigger() | Pauses a trigger so it stops creating work items. resume_trigger() starts it again. | Yes | No |
sdk.meta.work_items.list() | Lists work items across the Project or in one Workflow. get() reads one. | Yes | No |
sdk.meta.work_items.rename() | Renames a work item. | Yes | No |
sdk.meta.work_items.archive_many() | Archives up to 500 work items. unarchive_many() restores them. | Yes | No |
sdk.meta.work_items.rerun() | Starts a new run of a work item on its Workflow's latest version. | Yes | No |
sdk.meta.work_items.set_disposition() | Sets how a work item ended, for Workflows that declare outcomes. | Yes | No |
listWorkflows() | Lists the Project's Workflows. | No | Yes |
listTriggers() | Lists one Workflow's triggers. | No | Yes |
queryWorkItems() | Finds work items across every Workflow by status, metadata, outcome, or date. | No | Yes |
listWorkItems() | Lists one Workflow's work items. | No | Yes |
getWorkItem() | Reads one work item. | No | Yes |
getWorkItemAttachments() | Lists a work item's attachments. | No | Yes |
readWorkItemInputs() | Reads the inputs that work items were submitted with. | No | Yes |
readWorkItemOutputs() | Reads the outputs of several work items in one call. | No | Yes |
requestCreateWorkItem() | Opens the form that creates a work item. | No | Yes |
rerunWorkItem() | Starts a new run of a work item on its Workflow's latest version. | No | Yes |
archiveWorkItems() | Archives 1 to 20 work items, or restores them. | No | Yes |
openWorkItem() | Opens a work item in the app. | No | Yes |
startWorkItemConversation() | Opens a conversation about a work item. | No | Yes |
getModelUsage() | Reads the Project's model usage and estimated cost over a time range. | No | Yes |
Work with files
The agent reads and writes the Project's files directly in its workspace, so it needs no sdk call for ordinary files.
See Files.
| Call | What it does | Agent | View |
|---|---|---|---|
sdk.google.docs | Opens, creates, reads, and edits Google Docs that are in the Project. | Yes | No |
sdk.google.sheets | Opens, creates, reads, and writes Google Sheets that are in the Project. | Yes | No |
sdk.google.slides | Reads a summary of Google Slides that are in the Project, and replaces or updates their text. | Yes | No |
readJson() | Reads and parses a JSON file. | No | Yes |
listFiles() | Lists files under a folder. | No | Yes |
getFileUrl() | Returns a link to a file, for an image or a download. | No | Yes |
downloadFile() | Downloads a file to the person's computer. | No | Yes |
renderPdfPage() | Shows one page of a PDF as an image. | No | Yes |
openFile() | Opens a file in the app's file preview. | No | Yes |
Send email
The agent sends email from a shared mailbox or a hosted address configured for the workspace. See Email templates and Email-bound Workflows.
| Call | What it does | Agent | View |
|---|---|---|---|
sdk.mail.send() | Sends a new email. | Yes | No |
sdk.mail.reply() | Replies on the email thread that started the run. | Yes | No |
sdk.mail.update_inbound() | Marks the email that started the run as read, and adds or removes its categories or labels. | Yes | No |
sdk.mail.read_msg() | Reads an Outlook .msg file. | Yes | No |
Send notifications
Notifications go to people in the app, or to Slack or Microsoft Teams when the workspace connects them. See Notifications.
| Call | What it does | Agent | View |
|---|---|---|---|
sdk.notify.send() | Sends a notification, with optional rich Slack or Teams content. | Yes | No |
sdk.notify.digest() | Sends a compact digest notification. | Yes | No |
sdk.users.me() | Addresses the person the run works for. by_email() and by_id() address someone else in the workspace. | Yes | No |
| Helpers | sdk.notify.channel(), action(), and image() add a channel, a link button, or an annotated image. | Yes | No |
Use MCP servers and integrations
| Call | What it does | Agent | View |
|---|---|---|---|
sdk.mcp.list_servers() | Lists the MCP servers mounted in the Project. | Yes | No |
sdk.mcp.server() | Returns a server, to list and call its tools and read its resources and prompts. | Yes | No |
sdk.meta.mcp.install_catalog_server() | Installs a server from the catalog and mounts it. | Yes | No |
sdk.meta.mcp.mount() | Mounts an existing workspace server in the Project. | Yes | No |
sdk.meta.mcp.reauthorize() | Asks the person to sign in to a server again. | Yes | No |
sdk.meta.integrations.connect() | Connects an integration. enable() creates one, sync() refreshes it. | Yes | No |
sdk.meta.projects.attach_integration() | Attaches a workspace integration to a Project. | Yes | No |
sdk.meta.integrations.add_slack_conversations() | Adds Slack channels to the Slack integration. | Yes | No |
When an MCP server needs the person to sign in, the run asks them and continues after they do.
Use models and web search
| Call | What it does | Agent | View |
|---|---|---|---|
sdk.llm.prompt() | Sends one prompt to a model and returns the answer. | Yes | No |
sdk.llm.generate_images() | Generates images and saves them to the Project. | Yes | No |
sdk.llm.generate_image_batch() | Generates images for up to five prompts at once. | Yes | No |
sdk.exa.search() | Searches the web. | Yes | No |
sdk.exa.answer() | Answers a question from web results. | Yes | No |
sdk.exa.get_contents() | Reads the contents of web pages. | Yes | No |
sdk.exa.create_webset() | Builds a webset: a collection of web results that is kept up to date. | Yes | No |
The web search calls need the Exa integration.
Read documents
| Call | What it does | Agent | View |
|---|---|---|---|
sdk.docintel.parse() | Parses a document into structured content. | Yes | No |
sdk.docintel.extract() | Extracts fields from a document against a schema. | Yes | No |
sdk.docintel.infer_schema() | Drafts an extraction schema from a document and a description. | Yes | No |
sdk.docintel.wait_for_result() | Waits for a parse or an extraction to finish and returns its result. | Yes | No |
Show results in a conversation
| Call | What it does | Agent | View |
|---|---|---|---|
sdk.context.publish_vega_chart() | Shows an interactive chart in the conversation. | Yes | No |
sdk.context.add_image() | Adds an image for the agent to look at in its next step. add_vega_chart(), add_matplotlib_figure(), and add_pdf_pages() do the same for charts, figures, and PDF pages. | Yes | No |
sdk.views.preview() | Takes a screenshot of a custom view so the agent can check its own work. | Yes | No |
sdk.meta.threads.rename() | Renames the current conversation. | Yes | No |
sdk.tasks.list() | Lists the Project's background tasks. get() reads one. | Yes | No |
publishSelection() | Tells the conversation what the person selected in the view, so the agent can use it as context. | No | Yes |
Run Verified logic
| Call | What it does | Agent | View |
|---|---|---|---|
sdk.verified.list_approved() | Lists the approved Verified logic and the inputs each one takes. | Yes | No |
sdk.verified.run() | Runs approved Verified logic and returns its result. | Yes | No |
sdk.meta.verified_logic.run() | Runs the tests of a Verified logic package. | Yes | No |
listVerifiedLogic() | Lists the approved Verified logic. | No | Yes |
startVerifiedLogic() | Starts approved Verified logic with its inputs. | No | Yes |
getVerifiedLogicExecution() | Reads a run's state, result, and changed files. | No | Yes |
A view checks a run about once a second until it finishes. The view receives the result once; the platform does not save it. Files the logic writes are saved to the Project when the run succeeds.
Build and manage the Project
| Call | What it does | Agent | View |
|---|---|---|---|
sdk.meta.projects.create() | Creates a Project. open() opens one. | Yes | No |
sdk.meta.workflows.validate() | Checks a Workflow file before it is applied. | Yes | No |
sdk.meta.workflows.model_options() | Lists the models a Workflow's agent steps can use. | Yes | No |
sdk.meta.skills.validate() | Checks a skill package before it is published. | Yes | No |
sdk.meta.email_templates.validate() | Checks an email template. | Yes | No |
sdk.meta.evals | Lists evaluation case sets, runs them, and records the results. | Yes | No |
sdk.meta.tenant.list_projects() | Lists the Projects the person can see. list_integrations() and list_mcp_servers() do the same for integrations and MCP servers. | Yes | No |
sdk.meta.tenant.resolve_users() | Looks up people's names and pictures from their user IDs. | Yes | No |
What only a view does
| Call | What it does | Agent | View |
|---|---|---|---|
getContext() | Reads the Project, the person viewing, and what that person may do. | No | Yes |
getViewState() | Reads the person's saved state for this view. saveViewState() and clearViewState() change it. | No | Yes |
subscribeEvents() | Reacts to events, such as a file changing or a decision card closing. | No | Yes |
See What a view can read and do.
Permissions and limits
- The viewer's permissions apply. A view can do only what the person viewing it can do.
- Some view calls need operate access. Answering decisions, starting Verified logic, rerunning work items, and archiving work items need permission to operate the Project.
- Previews cannot act. A view in a preview cannot answer decisions or start Verified logic.
- Verified logic must be approved. A view can start only logic that a reviewer has approved.
- Effects are not undone. When a Verified logic run fails after it sent email or called a tool, those effects stay. Read the run's state before you start it again.
Ask the agent for the details. Each sdk module documents its own arguments.
Ask the agent in the Project's conversation how to use a call, and it reads the current
documentation from the module.