# Zerve User Documentation

## Docs

- [Welcome to Zerve](https://docs.zerve.ai/zerve-docs/readme.md): Zerve is an AI platform for data and research work, where agents plan, build, run and check the work for you.
- [How Zerve works](https://docs.zerve.ai/zerve-docs/getting-started/how-zerve-works.md): How Zerve's agents, notebooks and compute fit together.
- [Quickstart: your first superagent](https://docs.zerve.ai/zerve-docs/getting-started/quickstart-superagent.md): Create your first superagent, review its plan and watch it start work.
- [Quickstart: Evolve Mode in a notebook](https://docs.zerve.ai/zerve-docs/getting-started/quickstart-evolve.md): Run your first task with Evolve Mode, Zerve's most capable agent mode.
- [Key concepts](https://docs.zerve.ai/zerve-docs/getting-started/key-concepts.md): The terms used throughout the Zerve docs.
- [What is a superagent?](https://docs.zerve.ai/zerve-docs/superagents/superagents.md): A superagent is a long-running agent that owns a mission and manages a team of agents working toward it.
- [How the team works](https://docs.zerve.ai/zerve-docs/superagents/superagents/how-the-team-works.md): How a superagent's team of coordinators, workers and subagents is organized and works together.
- [How superagents approach work](https://docs.zerve.ai/zerve-docs/superagents/superagents/how-superagents-approach-work.md): The working principles superagents follow, from starting small to checking every result.
- [Create a superagent](https://docs.zerve.ai/zerve-docs/superagents/create-a-superagent.md): Create a superagent from a mission, review its plan and approve it to start work.
- [Objectives, KPIs and autonomy](https://docs.zerve.ai/zerve-docs/superagents/objectives-kpis-and-autonomy.md): Set what a superagent works toward, how success is measured and how freely it can decide.
- [Monitor a superagent](https://docs.zerve.ai/zerve-docs/superagents/monitor-a-superagent.md): Follow a superagent's progress, inspect its team's work and control it.
- [Decisions and recommendations](https://docs.zerve.ai/zerve-docs/superagents/decisions-and-recommendations.md): How a superagent asks you for decisions and inputs, and how to respond.
- [Conversations, backlog and follow-ups](https://docs.zerve.ai/zerve-docs/superagents/conversations-backlog-and-follow-ups.md): Talk to a superagent, see what it's working on and get updates when work finishes.
- [Batches: running work in parallel](https://docs.zerve.ai/zerve-docs/superagents/batches.md): How superagents run work in parallel, from scoring hundreds of inputs to judging competing approaches.
- [Map](https://docs.zerve.ai/zerve-docs/superagents/batches/map.md): Run the same task over many inputs in parallel.
- [Exploration](https://docs.zerve.ai/zerve-docs/superagents/batches/exploration.md): Try competing approaches scored on the same number, and keep the best.
- [Ideation](https://docs.zerve.ai/zerve-docs/superagents/batches/ideation.md): Brainstorm many distinct approaches, then rank them before committing to one.
- [Tournament](https://docs.zerve.ai/zerve-docs/superagents/batches/tournament.md): Pick the best of several finished pieces of work when quality is a judgement call.
- [Hill-climb](https://docs.zerve.ai/zerve-docs/superagents/batches/hill-climb.md): Improve one piece of work round after round, keeping what works.
- [Survey](https://docs.zerve.ai/zerve-docs/superagents/batches/survey.md): Run the same deep analysis across many entities, each with its own team.
- [How judging works](https://docs.zerve.ai/zerve-docs/superagents/batches/how-judging-works.md): How judges compare and rank work in Ideation, Tournament and judged Hill-climb batches.
- [Artifacts and outputs](https://docs.zerve.ai/zerve-docs/superagents/artifacts-and-outputs.md): The versioned files a superagent's team produces and builds on, and the outputs it produces for you.
- [Memory](https://docs.zerve.ai/zerve-docs/superagents/memory.md): How a superagent remembers decisions, findings and your preferences over time.
- [Schedules](https://docs.zerve.ai/zerve-docs/superagents/schedules.md): Have a superagent do work at set times, from a weekly briefing to a regular check on a model.
- [Channels](https://docs.zerve.ai/zerve-docs/superagents/channels.md): Reach a superagent from outside Zerve, through Slack, Telegram or X.
- [Slack](https://docs.zerve.ai/zerve-docs/superagents/channels/slack.md): Message a superagent in Slack.
- [Telegram](https://docs.zerve.ai/zerve-docs/superagents/channels/telegram.md): Message a superagent through a Telegram bot.
- [X](https://docs.zerve.ai/zerve-docs/superagents/channels/x.md): Let a superagent post, reply and search on X.
- [Superagent settings](https://docs.zerve.ai/zerve-docs/superagents/settings.md): Configure a superagent's identity, behavior, models, channels and budget.
- [Budgets and credits](https://docs.zerve.ai/zerve-docs/superagents/budgets-and-credits.md): How a superagent's daily budget works, and what happens when it runs out.
- [Best practices](https://docs.zerve.ai/zerve-docs/superagents/best-practices.md): How to get the best results from a superagent.
- [Threads](https://docs.zerve.ai/zerve-docs/agents-in-the-workspace/threads.md): Standalone conversations with Zerve's agent, for questions, research and analysis.
- [Agent modes](https://docs.zerve.ai/zerve-docs/agents-in-the-workspace/agent-modes.md): Choose how much Zerve's agent does on its own, from answering questions to running whole projects.
- [Evolve Mode](https://docs.zerve.ai/zerve-docs/agents-in-the-workspace/evolve-mode.md): Zerve's most capable agent mode, which plans, delegates to specialists, checks its work and learns as it goes.
- [Specialists](https://docs.zerve.ai/zerve-docs/agents-in-the-workspace/evolve-mode/specialists.md): The specialist agents Evolve Mode delegates work to.
- [When the agent needs input](https://docs.zerve.ai/zerve-docs/agents-in-the-workspace/evolve-mode/when-the-agent-needs-input.md): Choose whether Evolve Mode waits for your answers or carries on by itself.
- [Credit caps](https://docs.zerve.ai/zerve-docs/agents-in-the-workspace/evolve-mode/credit-caps.md): Limit how much a single Evolve Mode run can spend.
- [Learnings](https://docs.zerve.ai/zerve-docs/agents-in-the-workspace/learnings.md): The agent's persistent notes on your work, which carry over from one session to the next.
- [Sources and citations](https://docs.zerve.ai/zerve-docs/agents-in-the-workspace/sources-and-citations.md): The research references the agent saves and cites as it works.
- [Skills](https://docs.zerve.ai/zerve-docs/agents-in-the-workspace/skills.md): Reusable instructions and best practices the agent follows, built in or written by your team.
- [Custom subagents](https://docs.zerve.ai/zerve-docs/agents-in-the-workspace/custom-subagents.md): Define your own specialist agents for Evolve Mode to delegate to.
- [MCP servers](https://docs.zerve.ai/zerve-docs/agents-in-the-workspace/mcp-servers.md): Connect your own tools and data sources to Zerve's agent with MCP servers.
- [Models and effort](https://docs.zerve.ai/zerve-docs/agents-in-the-workspace/models-and-effort.md): How Zerve chooses which model the agent uses, and how to change a model's reasoning effort.
- [Supported models](https://docs.zerve.ai/zerve-docs/agents-in-the-workspace/supported-models.md): The model providers and models Zerve's agents can use.
- [Giving the agent context](https://docs.zerve.ai/zerve-docs/agents-in-the-workspace/giving-the-agent-context.md): Give the agent files, skills, other notebooks and data to work with.
- [Best practices](https://docs.zerve.ai/zerve-docs/agents-in-the-workspace/best-practices.md): How to get the best results from Zerve's agent in threads and notebooks.
- [Projects and folders](https://docs.zerve.ai/zerve-docs/notebooks/projects-and-folders.md): Organize your work into projects and folders, and find work shared with you.
- [The notebook and canvas](https://docs.zerve.ai/zerve-docs/notebooks/notebook-and-canvas.md): The two ways to view and work with a project in Zerve.
- [Layers](https://docs.zerve.ai/zerve-docs/notebooks/layers.md): Separate a canvas into workflows, such as development, deployment and scheduled jobs.
- [Blocks](https://docs.zerve.ai/zerve-docs/notebooks/blocks.md): The types of block you can use in a project, and what you can do with them.
- [Python](https://docs.zerve.ai/zerve-docs/notebooks/blocks/python.md): Run Python code in a block.
- [SQL query](https://docs.zerve.ai/zerve-docs/notebooks/blocks/sql-query.md): Query a data connection, or a dataframe from another block, with SQL.
- [R](https://docs.zerve.ai/zerve-docs/notebooks/blocks/r.md): Run R code in a block.
- [R Markdown](https://docs.zerve.ai/zerve-docs/notebooks/blocks/r-markdown.md): Combine R code and narrative, and publish it as a report.
- [Markdown](https://docs.zerve.ai/zerve-docs/notebooks/blocks/markdown.md): Add notes and documentation to a project.
- [GenAI](https://docs.zerve.ai/zerve-docs/notebooks/blocks/genai.md): Send a prompt to an AI model from a block, using variables from your project.
- [UI blocks (input and output)](https://docs.zerve.ai/zerve-docs/notebooks/blocks/ui-blocks.md): Build a simple app from your project with input and output blocks.
- [Aggregator](https://docs.zerve.ai/zerve-docs/notebooks/blocks/aggregator.md): Collect the results of a Fleet into one variable.
- [Logic gate](https://docs.zerve.ai/zerve-docs/notebooks/blocks/logic-gate.md): Run the next blocks only when a condition is met.
- [Variables, outputs and caching](https://docs.zerve.ai/zerve-docs/notebooks/variables-outputs-and-caching.md): How blocks share variables, how results are saved, and how to inspect them.
- [Language interoperability](https://docs.zerve.ai/zerve-docs/notebooks/language-interoperability.md): Pass data between Python, R and SQL blocks in the same project.
- [Files](https://docs.zerve.ai/zerve-docs/notebooks/files.md): Store, use and download files in a project.
- [Global imports](https://docs.zerve.ai/zerve-docs/notebooks/global-imports.md): Make imports available to every block in a project.
- [Data discovery](https://docs.zerve.ai/zerve-docs/notebooks/data-discovery.md): Have Zerve map out an unfamiliar database for you.
- [Block-level security](https://docs.zerve.ai/zerve-docs/notebooks/block-level-security.md): Lock blocks, and restrict who can see or change them.
- [Collaboration and comments](https://docs.zerve.ai/zerve-docs/notebooks/collaboration-and-comments.md): Work on projects with your team in real time, and discuss blocks with comments.
- [Keyboard shortcuts](https://docs.zerve.ai/zerve-docs/notebooks/keyboard-shortcuts.md): Keyboard shortcuts for the notebook and canvas.
- [How code runs](https://docs.zerve.ai/zerve-docs/compute/compute.md): How Zerve runs your code, block by block, on the compute you choose.
- [Compute types](https://docs.zerve.ai/zerve-docs/compute/compute-types.md): The kinds of compute a block can run on, and when to use each.
- [GPUs](https://docs.zerve.ai/zerve-docs/compute/gpus.md): Run blocks on GPUs, and what to do when GPUs are in short supply.
- [Fleet: parallel runs](https://docs.zerve.ai/zerve-docs/compute/fleet.md): Run the same block many times in parallel with Fleet.
- [Limits and concurrency](https://docs.zerve.ai/zerve-docs/compute/limits-and-concurrency.md): The limits on block runs, and how many can run at once.
- [Troubleshooting block runs](https://docs.zerve.ai/zerve-docs/compute/troubleshooting-block-runs.md): What common block run errors mean, and how to fix them.
- [Environments and packages](https://docs.zerve.ai/zerve-docs/environments/environments.md): The packages and settings your code runs with, and how to add your own.
- [Environment presets](https://docs.zerve.ai/zerve-docs/environments/environment-presets.md): Ready-made environments for common kinds of work in Python and R.
- [Executor images](https://docs.zerve.ai/zerve-docs/environments/executor-images.md): Build and manage your organization's own environments.
- [Assets overview](https://docs.zerve.ai/zerve-docs/data-and-assets/data.md): Reusable data connections, constants, secrets and functions, shared across notebooks.
- [Data connections](https://docs.zerve.ai/zerve-docs/data-and-assets/connections.md): Connect Zerve to your databases and data platforms.
- [PostgreSQL](https://docs.zerve.ai/zerve-docs/data-and-assets/connections/postgresql.md): Connect to a PostgreSQL database.
- [MySQL](https://docs.zerve.ai/zerve-docs/data-and-assets/connections/mysql.md): Connect to a MySQL database.
- [MariaDB](https://docs.zerve.ai/zerve-docs/data-and-assets/connections/mariadb.md): Connect to a MariaDB database.
- [Snowflake](https://docs.zerve.ai/zerve-docs/data-and-assets/connections/snowflake.md): Connect to Snowflake, with a password or key-pair authentication.
- [Databricks](https://docs.zerve.ai/zerve-docs/data-and-assets/connections/databricks.md): Connect to Databricks with an access token.
- [BigQuery](https://docs.zerve.ai/zerve-docs/data-and-assets/connections/bigquery.md): Connect to Google BigQuery with a service account key.
- [KX (kdb+)](https://docs.zerve.ai/zerve-docs/data-and-assets/connections/kx.md): Connect to a kdb+ database from KX.
- [Ocient](https://docs.zerve.ai/zerve-docs/data-and-assets/connections/ocient.md): Connect to an Ocient database.
- [Weaviate](https://docs.zerve.ai/zerve-docs/data-and-assets/connections/weaviate.md): Connect to a Weaviate vector database.
- [Domo](https://docs.zerve.ai/zerve-docs/data-and-assets/connections/domo.md): Connect to Domo with a client ID and secret.
- [Constants and secrets](https://docs.zerve.ai/zerve-docs/data-and-assets/constants-and-secrets.md): Store named values, such as API keys and configuration, and use them across notebooks.
- [Functions](https://docs.zerve.ai/zerve-docs/data-and-assets/functions.md): Share reusable code across notebooks, with versions.
- [DAGs](https://docs.zerve.ai/zerve-docs/data-and-assets/dags.md): Save connected blocks as a reusable pipeline, and reuse it across projects.
- [Deployments overview](https://docs.zerve.ai/zerve-docs/ship-your-work/shipping.md): Turn your work into apps, APIs, scheduled jobs and reports.
- [Apps](https://docs.zerve.ai/zerve-docs/ship-your-work/apps.md): Deploy an interactive web app, such as a dashboard, from your project.
- [APIs](https://docs.zerve.ai/zerve-docs/ship-your-work/apis.md): Deploy your model or code as an API that other systems can call.
- [Scheduled jobs](https://docs.zerve.ai/zerve-docs/ship-your-work/scheduled-jobs.md): Run a project automatically on a schedule.
- [Reports](https://docs.zerve.ai/zerve-docs/ship-your-work/reports.md): Turn a notebook into an interactive report your team can ask questions of.
- [Git integration](https://docs.zerve.ai/zerve-docs/git-and-version-control/git.md): Connect Zerve to your Git provider to version your projects.
- [GitHub (cloud)](https://docs.zerve.ai/zerve-docs/git-and-version-control/git/github-cloud.md): Connect Zerve to your GitHub repositories.
- [GitHub Enterprise Server](https://docs.zerve.ai/zerve-docs/git-and-version-control/git/github-enterprise-server.md): Connect Zerve to your GitHub Enterprise Server.
- [Bitbucket OAuth consumer](https://docs.zerve.ai/zerve-docs/git-and-version-control/git/bitbucket-oauth-consumer.md): Connect Zerve to Bitbucket with access tied to a user account.
- [Bitbucket Cloud Connect app](https://docs.zerve.ai/zerve-docs/git-and-version-control/git/bitbucket-cloud-connect-app.md): Connect Zerve to all the repositories in a Bitbucket workspace.
- [Azure DevOps Repos](https://docs.zerve.ai/zerve-docs/git-and-version-control/git/azure-devops-repos.md): Connect Zerve to your Azure DevOps repositories.
- [Source control in notebooks](https://docs.zerve.ai/zerve-docs/git-and-version-control/source-control-in-notebooks.md): Commit, branch and restore your projects with Git, and use code from your repositories.
- [Organizations](https://docs.zerve.ai/zerve-docs/administration/administration.md): How organizations work in Zerve, and where to manage yours.
- [Members and roles](https://docs.zerve.ai/zerve-docs/administration/members-and-roles.md): Invite people to your organization and control what they can do.
- [Sharing and privacy](https://docs.zerve.ai/zerve-docs/administration/sharing-and-privacy.md): Control who can see and edit your work.
- [Plans, credits and billing](https://docs.zerve.ai/zerve-docs/administration/plans-credits-and-billing.md): Zerve's plans, how credits work, and what happens when they run out.
- [How credits are used](https://docs.zerve.ai/zerve-docs/administration/how-credits-are-used.md): What uses credits in Zerve, how much models cost, and how to keep spending in check.
- [Usage monitoring](https://docs.zerve.ai/zerve-docs/administration/usage-monitoring.md): See how your organization uses credits and compute.
- [SAML single sign-on](https://docs.zerve.ai/zerve-docs/administration/saml-sso.md): Set up SAML single sign-on so your team signs in to Zerve through your identity provider.
- [Model settings](https://docs.zerve.ai/zerve-docs/administration/model-settings.md): Choose which AI providers and models people in your organization can use.
- [Bring your own API keys](https://docs.zerve.ai/zerve-docs/administration/bring-your-own-api-keys.md): Use your organization's own model provider accounts with Zerve.
- [Organization settings](https://docs.zerve.ai/zerve-docs/administration/organization-settings.md): Organization-wide settings for admins.
- [Preferences and notifications](https://docs.zerve.ai/zerve-docs/administration/preferences-and-notifications.md): Your personal settings for the editor, the agent and notifications.
- [Developer API](https://docs.zerve.ai/zerve-docs/administration/developer-api.md): Create API keys for working with Zerve programmatically.
- [Deployment options](https://docs.zerve.ai/zerve-docs/self-hosting/self-hosting.md): The ways you can run Zerve, from Zerve's cloud to your own cloud account or your own infrastructure.
- [AWS QuickStart](https://docs.zerve.ai/zerve-docs/self-hosting/aws-quickstart.md): Run Zerve's compute and storage in your own AWS account, set up with a guided CloudFormation stack.
- [What Zerve can access in your AWS account](https://docs.zerve.ai/zerve-docs/self-hosting/aws-access.md): Exactly what access Zerve has in your AWS account when you use bring your own cloud, for security reviews.
- [AWS CloudFormation reference](https://docs.zerve.ai/zerve-docs/self-hosting/aws-cloudformation-reference.md): Everything the AWS QuickStart's CloudFormation stack creates in your account.
- [GCP (GKE)](https://docs.zerve.ai/zerve-docs/self-hosting/gcp-gke.md): Run Zerve's compute and storage in your own Google Cloud project, on your existing GKE cluster.
- [Azure (AKS)](https://docs.zerve.ai/zerve-docs/self-hosting/azure-aks.md): Run Zerve's compute and storage in your own Azure account, on your existing AKS cluster.
- [Self-managed installs](https://docs.zerve.ai/zerve-docs/self-hosting/self-managed.md): Run the whole of Zerve on your own infrastructure, including on-premises and air-gapped installs.
- [Security overview](https://docs.zerve.ai/zerve-docs/security/security.md): How Zerve keeps your data, code and credentials secure.
- [Agent approvals and governance](https://docs.zerve.ai/zerve-docs/security/agent-approvals-and-governance.md): How Zerve's agents ask for approval, and the controls you have over what they do.
- [Limits](https://docs.zerve.ai/zerve-docs/reference/limits.md): Every limit in Zerve, in one place.
- [Terms of Service](https://docs.zerve.ai/zerve-docs/reference/terms-of-service.md)
- [Privacy Policy](https://docs.zerve.ai/zerve-docs/reference/privacy-policy.md)
- [End User License Agreement](https://docs.zerve.ai/zerve-docs/reference/end-user-license-agreement.md)


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
This site has an agentic ask interface you may use to query the documentation dynamically by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.zerve.ai/zerve-docs/index.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
