# AIOps Control Deck documentation > AIOps Control Deck is a multi-tenant AIOps platform. It ingests alerts and cloud scanner findings, correlates them into incidents and hands each incident to an incident, change or compliance agent. The agents investigate through read-only connections, write a root-cause analysis and propose fixes. A fix runs after people approve it; only a pre-approved route, in a workspace that switched automatic fixes on, runs one without that step. This index covers release 0.1.164. People and agents should start with Quick start. Links are absolute paths on this installation; every page is also available as HTML at the same path without `.md`, and `Accept: text/markdown` on that path returns the Markdown. Rules for an AI agent that acts on a workspace: - Make read-only calls first: `GET /api/aiops/me`, then `GET /api/aiops//datasources` and `.../incidents`. - A person mints your first token in the product, under **Settings · Data in → Alert intake & tokens → New API token**. Ask for the narrowest scope that works. Never write a token to a file. - Never try to bypass approval: releasing a fix needs a person with `approvals.decide`, and an L3 fix needs two different people. ## Quick start - [1. Your first investigation](/help/getting-started.md): Connect one read source, discover its context, send a test alert and follow the answer. Add approved fixes… - [2. Send your first alert](/help/send-first-alert.md): Create a test alert, send its sample with one curl command and open the resulting incident. - [3. Rehearse your first approved fix](/help/first-approved-fix.md): Open a test security-group rule, let the agent prepare its fix, release it through one approval, roll it back… ## Get started - [Get started index](/help/get-started/llms.txt): all 13 Get started pages with their summaries, 9 more in the index - [What the platform does](/help/what-the-platform-does.md): The seven agents, the path from an alert to a verified fix, connections and roles, what spends model budget… - [Start here: your first investigation](/help/getting-started.md): Connect one read source, discover its context, send a test alert and follow the answer. Add approved fixes… - [Send your first alert](/help/send-first-alert.md): Create a test alert, send its sample with one curl command and open the resulting incident. - [Rehearse your first approved fix](/help/first-approved-fix.md): Open a test security-group rule, let the agent prepare its fix, release it through one approval, roll it back… ## Install and operate the platform - [Install and operate the platform index](/help/install-and-operate-the-platform/llms.txt): all 21 Install and operate the platform pages with their summaries, 17 more in the index - [Architecture and responsibilities](/help/ops-architecture-and-responsibilities.md): This runbook covers the managed installation in a dedicated customer AWS account. - [Install](/help/ops-install.md): The vendor installs the platform into your dedicated AWS account from its deployment pipeline, in a window… - [Customer deploy runbook — a fresh AWS account in 8 steps](/help/ops-installation-runbook.md): The managed install into one dedicated AWS account in eight steps, with upgrade, rollback, teardown, costs… - [Licence](/help/ops-licence.md): The licence is an offline signed file verified on the control plane at boot and scheduler ticks. It makes no… ## Connect your environment - [Connect your environment index](/help/connect-your-environment/llms.txt): all 57 Connect your environment pages with their summaries, 53 more in the index - [Choose your connection](/help/choose-your-connection.md): Alert source, read connection, write credential, private-network agent or ticket system — what each agent… - [Connect AWS](/help/setup-aws.md): Review a CloudShell setup command, import its public output and verify separate AWS connections. - [Proactive monitoring](/help/proactive-monitoring.md): Incidents without alerts of your own — the platform watches endpoints, metrics and the cloud's own state, and… - [Built-in detection (AWS)](/help/builtin-detection.md): CloudWatch rules over your estate inventory for accounts with no alarms — outages, and latency or faults… ## Work with the agents - [Work with the agents index](/help/work-with-the-agents/llms.txt): all 32 Work with the agents pages with their summaries, 28 more in the index - [Dashboard](/help/dashboard.md): The overall dashboard — pillar rings around the human gate, service KPIs, pillar health, the gate queue, one… - [Dashboard, Operate and alerts](/help/mission-overview.md): Read your next action on the Dashboard, follow the work under Operate, and trace an incoming alert to its… - [Incidents](/help/incidents.md): The incident is the unit of work. Where to read how one opens, what its page shows, how a fix is approved… - [How incidents work](/help/how-incidents-work.md): The concepts, the six statuses and the moves between them, severity, priority and risk tier, and how the… ## Administer a workspace - [Administer a workspace index](/help/administer-a-workspace/llms.txt): all 11 Administer a workspace pages with their summaries, 7 more in the index - [People, access and audit](/help/access-and-audit.md): Where membership and roles, single sign-on, support access and the audit log are documented — one page each —… - [Members and roles](/help/members-and-roles.md): Invite, re-role and remove members, create custom roles, what every workspace includes, the tenant brand, and… - [Single sign-on](/help/single-sign-on.md): Register this platform at your identity provider, connect it, map groups to roles, test discovery, what login… - [Support access](/help/support-access.md): The expiring, audited operator grant platform staff use when you ask for help, how you end it, and exactly… ## Reference - [Reference index](/help/reference/llms.txt): all 18 Reference pages with their summaries, 14 more in the index - [Integrations: MCP, API, webhooks, CLI](/help/integrations.md): The hub for everything outside the browser: the REST API, personal access tokens, the MCP server, the agent… - [REST API](/help/rest-api.md): The concepts every doorway shares, your tenant id, the live OpenAPI schema, the endpoints most people want… - [Personal access tokens](/help/personal-access-tokens.md): Create a personal access token, choose its scope and lifetime, and rotate or revoke it; the credential the… - [Agent API (OpenAI-compatible)](/help/agent-api.md): Each pillar agent as an OpenAI-compatible chat-completions endpoint: live and sandbox base URLs, curl and… ## Troubleshooting - [Troubleshooting index](/help/troubleshooting/llms.txt): all 2 Troubleshooting pages with their summaries - [Troubleshooting hub](/help/troubleshooting.md): Find the failed step and the evidence needed to repair it. - [What to collect before you escalate](/help/before-you-escalate.md): The ids, run header, failed tool calls, connection test and audit entries to gather before you ask for help. ## What's new - [What's new index](/help/what-s-new/llms.txt): all 20 What's new pages with their summaries, 16 more in the index - [What's new](/help/whats-new.md): The customer-visible changes in 0.1.146 to 0.1.164 and in 0.1.130 to 0.1.133 — what changed on your screens… - [Release notes by version](/help/release-notes.md): Customer-visible changes, newest first. The current released version covered here is 0.1.164; GET /healthz… - [Release notes 0.1.164](/help/release-notes-0-1-164.md): 28 October 2026: A certified runbook demoted for a fix that did not hold, before the outside write that… - [Release notes 0.1.163](/help/release-notes-0-1-163.md): 28 October 2026: An outside change and the scanner finding that followed it now meet on the finding. ## API - [OpenAPI spec](/openapi.json): the live schema of this installation - [REST API](/help/rest-api.md): tenant id, the live schema, pagination, live investigation stream - [Personal access tokens](/help/personal-access-tokens.md): create a token, scopes, lifetime, rotation - [Agent API (OpenAI-compatible)](/help/agent-api.md): one completion = one investigation; live and sandbox flavours - [Webhook reference](/help/webhook-reference.md): every intake endpoint, the token forms, payloads, counters and errors ## MCP - [MCP server (tenant)](/help/mcp-server.md): /mcp/aiops endpoint, authentication, client setup, operations - [MCP operation reference](/help/mcp-operations.md): every tool and its REST operation - [Admin MCP server](/help/ops-admin-mcp.md): /mcp/admin for platform staff ## Optional - [Full text for offline indexing](/help/llms-full.txt): every page in one file; too big for one fetch, never the entry point