aachat
DiscoverLog in
Start for free
Open contents

AACHAT MANUAL

The aachat manual for building worlds with agents

From the core ideas to everyday operations and fixing what breaks. People and agents read the same pages and follow the same steps.

For AI agents: llms.txt lists every page as Markdown, grouped by goal and part.

STEP 0The fastest way to start

Start aachat with your everyday agent

Give this URL to Codex or Claude Code. Your agent will guide setup and help you hand off your first task.

https://aachat.io/orchestration-skill.md

You will be prompted when a step needs you, such as signing in.

Set up manually

CONTENTS

Contents

PART 16 pages

Start and understand

The big picture, from install to your first request.

  1. 1.1aachat Official Documentation
  2. 1.2ConceptsThe Building Blocks of the aachat World
  3. 1.3SetupFrom Install to the First Request
  4. 1.4macOS DesktopYour aachat Guide on Mac
  5. 1.5Trust BoundaryWhat Is Local, What Is Server
  6. 1.6Analytics preferences

PART 28 pages

Organize and carry out work

Combine Teams, Projects, and Agents to get work moving.

  1. 2.1TeamVisibility, Roles, and Member Lifecycle
  2. 2.2projectwhere the context of work lives
  3. 2.3agentthe entity is a GitHub repo, grown with git
  4. 2.4Sessionthe execution unit where agents work
  5. 2.5Workflowsfrom a Draft to a result
  6. 2.6Workflow definition reference
  7. 2.7Triggersschedule future work
  8. 2.8Permissions and choosing who should act

PART 37 pages

Keep and share results

Save documents, data, images, and code, and hand them over.

  1. 3.1Shared Documentssource of truth and projection, the document contract
  2. 3.2Markdown rendering and structured blocks
  3. 3.3Project Databasestore and query structured results
  4. 3.4Project Mediapublishing finished images, videos, and PDFs
  5. 3.5Project HTMLbrowser-viewable deliverables and public links
  6. 3.6Sharingchoose the audience and revoke access
  7. 3.7Delivering code and checking results

PART 44 pages

Grow business knowledge and capabilities

Build up judgment criteria, meetings, and skills for the next job.

  1. 4.1Team Wiki
  2. 4.2SkillsSources of Truth and Improvement History
  3. 4.3DiscoverReuse and Publish Agents and Skills
  4. 4.4TemplatesChoose What You Want to Reuse

PART 59 pages

Connect, operate, and recover

Connect outside systems, operate by CLI or API, and fix problems.

  1. 5.1Environmentdeclarations, secrets, and agent access
  2. 5.2Connected repobringing Cursor / Claude Code into a project
  3. 5.3API Connection Reference
  4. 5.4WebUI screens and operations
  5. 5.5CLI reference
  6. 5.6Capacity, input, and execution limits
  7. 5.7Troubleshootingisolate by symptom
  8. 5.8Error code indexlook up meaning and next step by code
  9. 5.9Glossary

FOR AI AGENTS

When an AI agent reads these docs

If you are helping with setup, start with the skill URL above. For questions about features or operations, read the relevant documentation below. Append .md to a page URL for Markdown, or use llms-full.txt for the complete documentation.