# Industric

> Industric is a control plane for AI agents, workflows, and governance across managed deployments, local devboxes, and supported workstations. Build agents, connect tools and MCP servers, share knowledge through AgentWiki and inboxes, and inspect recorded sessions and policy decisions.

- Docs are served from https://exo-docs.adaptive.live. Every page below is also readable as HTML at its URL.
- `llms-full.txt` carries the full prose of the same pages in one file.
- Kubernetes resources use the `autonomic.sh/v1alpha1` API group. The CLI is `exo`.

## Getting started

- [Documentation](https://exo-docs.adaptive.live/): Set up Industric, run your first agent, and manage the tools and policies it uses.
- [Architecture](https://exo-docs.adaptive.live/start/architecture)
- [Core concepts](https://exo-docs.adaptive.live/start/concepts)
- [Getting started](https://exo-docs.adaptive.live/start/getting-started): Get a platform you can reach and put the exo CLI on your machine. Then go create your first agent.
- [Create your first Agent](https://exo-docs.adaptive.live/start/create-first-agent): Configure a provider, connect a Tool, create an Agent, and run your first session.

## Deploy & operate

- [Prerequisites & topology](https://exo-docs.adaptive.live/deploy/prerequisites): What you're installing, where it runs, and what the cluster needs before you apply anything.
- [Managed Cloud](https://exo-docs.adaptive.live/deploy/managed/setup): On managed cloud, Industric runs the control plane for you. The only thing you install is the operator in your own cluster, then point it at managed cloud with a deployment token.
- [Agent Cloud](https://exo-docs.adaptive.live/deploy/agent-cloud): An opt-in deploy mode where Industric provisions a Kubernetes cluster for you and installs the operator into it automatically, so a deployment goes from created to running agents with no manual setup.
- [Resource requirements](https://exo-docs.adaptive.live/deploy/resource-planning): CPU, memory, and storage for Exo workloads.
- [High availability](https://exo-docs.adaptive.live/deploy/high-availability): Keeping your Exo workloads available and knowing when something needs attention.
- [Disaster recovery](https://exo-docs.adaptive.live/deploy/disaster-recovery): Recovering from a major outage with warm standby.

## Connect & work

- [Tool](https://exo-docs.adaptive.live/connect/tools): Connect to a database, cluster, host, or API through a managed connection. Credentials are supplied to the tool so you do not need to type them into commands. Tools can still expose them in output.
- [Chat](https://exo-docs.adaptive.live/connect/chat): Work with models, tools, skills, and agents in the platform, and continue the same conversations from the CLI.
- [Agent](https://exo-docs.adaptive.live/connect/agents): Connect to an agent, run a task, share a live session, or start from a template. Access is governed and sessions are recorded.
- [Webhooks](https://exo-docs.adaptive.live/connect/agent-webhooks)

## Govern & observe

- [Access control](https://exo-docs.adaptive.live/govern/access-control): Control who can view, edit, or connect to agents.
- [Guardrails](https://exo-docs.adaptive.live/govern/guardrails): Give agents standing prompt instructions, scoped to the people and agents they apply to.
- [Classifiers](https://exo-docs.adaptive.live/govern/classifiers): Detect sensitive content with built-in classifiers, custom rules, and decision models.
- [OPA integration](https://exo-docs.adaptive.live/govern/opa): Connect your Open Policy Agent server to control tool runs, VM and Kubernetes shells, MCP delivery, and LLM gateway traffic. Your policies stay on your network.
- [Resource policies](https://exo-docs.adaptive.live/govern/typed-policies): Define tool, MCP, command, and gateway policies, group them into policy bundles, and control which resources they cover. Test rules, stage rollout, and review approvals and decisions.
- [Usage](https://exo-docs.adaptive.live/govern/usage): See your usage and limits for LLM credits, disk space, and Agent Cloud nodes.
- [Sessions](https://exo-docs.adaptive.live/observe/sessions): The full record of an agent run: tool calls, checkpoints, and terminal I/O, captured live and replayable frame by frame with the original timing.
- [Audit](https://exo-docs.adaptive.live/observe/audit): The log of who did what on the platform (logins, agent connects, resource changes) and what happened in all agent sessions, searchable by actor, kind, and time.

## Developers

- [Industric harness](https://exo-docs.adaptive.live/build/harness): Run HAL in a managed container or on your machine, with shared chat, history, and approvals.
- [CLI](https://exo-docs.adaptive.live/cli/overview): Run cluster agents and tools, manage dashboard resources, and sync skills from the terminal.
- [Multi-Agent](https://exo-docs.adaptive.live/connect/terminal-dock): Keep terminals running as you browse, refresh, or work with several agents at once.
- [Agent inbox](https://exo-docs.adaptive.live/connect/inbox): Read agent reports, send messages, and start runs from the inbox.
- [Model providers](https://exo-docs.adaptive.live/build/providers): Set up your AI providers and API keys once for your agents to use.
- [Runtimes](https://exo-docs.adaptive.live/build/runtimes): Choose the software that runs your agent and connect it to a model provider.
- [Agent](https://exo-docs.adaptive.live/build/agents): Create an agent with a runtime, goal, provider, skills, and schedule.
- [Workflows](https://exo-docs.adaptive.live/build/workflows): Connect agents into workflows with triggers, approvals, and run history.
- [Schedules](https://exo-docs.adaptive.live/build/schedules): Run agents on a schedule or once at a set time.
- [Tool](https://exo-docs.adaptive.live/build/tools): Connect agents and people to databases, servers, APIs, and Kubernetes.
- [Skills](https://exo-docs.adaptive.live/build/skills): Write reusable instructions for agents and sync them to your machine.
- [MCP servers](https://exo-docs.adaptive.live/build/mcp): Connect MCP servers, share their tools with agents, and track usage.
- [Memory](https://exo-docs.adaptive.live/build/memory): Save knowledge for agents to reuse and update across runs.
- [File systems](https://exo-docs.adaptive.live/build/storage): Store and share agent files across restarts. Browse them in the dashboard or transfer them with the CLI.
- [LLM gateways](https://exo-docs.adaptive.live/build/gateways): Manage model access, track usage, and set budgets for agents and users.
- [Apps](https://exo-docs.adaptive.live/build/apps): Build apps with a coding agent, preview them live, and add Postgres when needed.
- [Web agents](https://exo-docs.adaptive.live/build/web-agents): Give agents a browser you can watch and control from the dashboard.

## Agent Sentinel

- [Agent Sentinel](https://exo-docs.adaptive.live/govern/sentinel): See which AI agents run on your workstations, what they do, and apply supported controls to them with the separate exo-sentinel service.
- [Deploy Sentinel to devices](https://exo-docs.adaptive.live/govern/sentinel/deploy): Create a company deployment, push it through your device management tool, and roll it out ring by ring.
- [Activity & alerts](https://exo-docs.adaptive.live/govern/sentinel/activity): Decide how much prompt content Sentinel collects, then search activity across devices, see who is running agents, and review model-assessed alerts.
- [Network controls](https://exo-docs.adaptive.live/govern/sentinel/network): Capture and control workstation traffic to AI vendors, and limit which applications can reach AI services.
- [Gateway routing](https://exo-docs.adaptive.live/govern/sentinel/gateway): Route provider calls from workstations through an LLM gateway, with keys bound to the person and process.
- [Supervised runs](https://exo-docs.adaptive.live/govern/sentinel/supervision): Start workstation agents through Sentinel so they run in a sandbox with credential and network protections, and decide what happens to agents started any other way.
- [Inventory & add-on policy](https://exo-docs.adaptive.live/govern/sentinel/inventory): See the MCP servers, skills, add-ons, and IDE plugins configured on each device, and block the add-ons you do not allow.

## Identity & access

- [Auth model overview](https://exo-docs.adaptive.live/identity/auth-model): Which credential to use for people, automation, deployments, and agents.
- [Your account](https://exo-docs.adaptive.live/identity/personal-settings): The self-service account page every user has: view your profile, mint a short-lived CLI access token, and enroll two-factor authentication. No admin rights needed.
- [Roles & permissions](https://exo-docs.adaptive.live/identity/roles): A permission catalog, two seeded roles, tenant-authored roles and permissions, and the difference between a Team and an IdP Group.
- [SSO · SAML & OIDC](https://exo-docs.adaptive.live/identity/sso): Sign in to Industric through your identity provider. Register Industric as an app in the IdP, enter the provider values in the dashboard, and the browser sign-in flow mints the same User JWT a password login would.
- [Directory sync](https://exo-docs.adaptive.live/identity/directory-sync): Pull users and groups into Industric from your directory on a schedule. Industric connects out to LDAP, Microsoft Entra, or Google Workspace, reconciles people and groups, and keeps memberships current between logins.
- [SCIM provisioning](https://exo-docs.adaptive.live/identity/scim): Let your identity provider push users and groups into Industric in real time. Industric runs a standard SCIM 2.0 server. Point your IdP at the base URL, paste in a bearer token, and provisioning, updates, and deactivations flow automatically.

## Reference

- [Glossary](https://exo-docs.adaptive.live/resources/glossary)
- [CRD overview](https://exo-docs.adaptive.live/resources/crd-overview): Everything Industric runs in your cluster is a Kubernetes custom resource under autonomic.sh/v1alpha1. What the kinds are, how to install the definitions, and how to author and inspect objects.
- [LLMGateway](https://exo-docs.adaptive.live/resources/llmgateway): The infrastructure CRD behind an LLM gateway: the operator reconciles it into a StatefulSet plus a Service that agents in the namespace point at.
- [Troubleshooting & FAQ](https://exo-docs.adaptive.live/resources/troubleshooting): The issues people actually hit, with the exact error text, what it means, and the fix. Searchable by error message.
- [Changelog](https://exo-docs.adaptive.live/changelog): Product updates and changes to the documentation.

## Optional

- [Connecting to the control plane](https://exo-docs.adaptive.live/autonomic/connecting): Turn on managed mode: mint a deployment token, mount it as a Secret, and point the manager at your platform over a single outbound WebSocket.
- [Deploying the manager](https://exo-docs.adaptive.live/autonomic/deploying-manager): The autonomic-manager operator: its namespace, the ServiceAccount and ClusterRole it needs, the Service it exposes, and the admission webhook it self-bootstraps.
- [Installing Industric Operator](https://exo-docs.adaptive.live/autonomic/install): Operator install moved into the Deploy chapter and the Industric Operator pages.
- [What is Industric Operator](https://exo-docs.adaptive.live/autonomic/overview): The in-cluster operator that reconciles the autonomic.sh CRDs into running workloads and connects your deployment to the Industric control plane over one outbound WebSocket.
- [Runtime images](https://exo-docs.adaptive.live/autonomic/runtime-images): The operator doesn't run your agents itself. It launches pods from a set of adaptivelive/* images. Common images and how to change them. Check your release inventory for the full set.
- [Tokens & deployments](https://exo-docs.adaptive.live/autonomic/tokens): The lifecycle of a deployment token: minting, rotating, and the rule that keeps two clusters from sharing one identity.
- [Code Review & Composer](https://exo-docs.adaptive.live/build/code-review): Review pull requests automatically or open a cluster devbox to prepare a fix pull request.
- [Devboxes](https://exo-docs.adaptive.live/build/devboxes): Run a repository with its toolchains and coding harness in local Docker or a cluster, with recorded sessions and a persistent workspace.
- [Testboxes](https://exo-docs.adaptive.live/build/testboxes): Turn a plain-language test plan into repeatable UI, smoke, integration, or unit tests with retained run artifacts.
- [AgentWiki](https://exo-docs.adaptive.live/build/wiki): Build a searchable knowledge base from your environment and keep operator and agent notes across inventory refreshes.
- [Skill Sync](https://exo-docs.adaptive.live/connect/skill-sync): Pull the skill catalogue onto your laptop with exo skill sync. A skill authored in Industric lands in the on-disk layout your local coding agent expects — no copy-paste, hand-authored files left untouched.
- [Air-gapped & private registries](https://exo-docs.adaptive.live/deploy/air-gapped): Run Industric from a private registry with no public-internet access. Mirror the required images and configure local services and dependencies for offline use.
- [Control-plane config & secrets](https://exo-docs.adaptive.live/deploy/config-secrets): Basic backend settings, required secrets, and production configuration reminders.
- [Database · Postgres](https://exo-docs.adaptive.live/deploy/database): Industric's only stateful dependency. What version, how the schema gets built and maintained, and the options for storing session recordings.
- [Install with deploy.yaml](https://exo-docs.adaptive.live/deploy/deploy-yaml): The raw manifest path for the control plane. Use this when you need a plain YAML file under GitOps. Use exo-install for the control plane, operator, and CRDs, with a separate database migration step.
- [Email · SMTP & Mailgun](https://exo-docs.adaptive.live/deploy/email): Configure outbound email so password resets and invitations actually send. Pick SMTP or Mailgun; without one, both flows fail silently.
- [Install with exo-install](https://exo-docs.adaptive.live/deploy/exo-install): The recommended path. One binary applies the CRDs, the operator, and the platform through your own kubectl, with pre-flight checks and generated secrets.
- [Managed Cloud · Upgrades & versioning](https://exo-docs.adaptive.live/deploy/managed/upgrades): On managed cloud the platform is upgraded for you. The only piece you keep current is the operator in your own cluster.
- [Manual install · manifests & Helm](https://exo-docs.adaptive.live/deploy/manifests): Raw manifests under GitOps instead of the installer: the platform deploy.yaml, the Helm chart, the CRD bundle, and the edits the stock files need.
- [TLS, ingress & networking](https://exo-docs.adaptive.live/deploy/networking): The backend serves plain HTTP and assumes you terminate TLS at the edge. What to expose, the WebSocket paths that must survive your proxy, and which way traffic flows between clusters.
- [Upgrades & versioning](https://exo-docs.adaptive.live/deploy/upgrades): The platform and the operator version independently. How to roll each one forward, the one component Helm won't upgrade for you, and how to back out.
- [LLM Gateway](https://exo-docs.adaptive.live/govern/llm-gateway): An optional proxy Industric deploys into your own cluster. It fronts your provider keys, mints short-lived scoped virtual keys for agent runs, and meters every request against spend budgets.