AOS Work Playbook
This is the AOS Work playbook. It skips model theory and stays on how you use it in your actual work: where to start with a task, which resources it needs, and where the output gets checked.
The four chapters cover four layers: meeting the product and its resources, turning one result into a system, single-task office scenarios, and role scenarios. The lessons follow a dependency order, and every section number and cross-reference in the text matches it.
What Each Chapter Covers
| Chapter | Coverage | The question it answers |
|---|---|---|
| Chapter 1 | Product positioning, the interface, the four resource types, the two context mechanisms | How does it work, and what do I need to set up |
| Chapter 2 | Workspaces and automation, building Skills, designing Work, taking assets further | How do I make this work for me over the long term |
| Chapter 3 | Six office tasks, end to end | How do I do this task in front of me |
| Chapter 4 | Eight role scenarios, end to end | What are the established uses for my role |
How One Task Actually Runs
The whole book is about one thing: you hand the work over, it hands back the finished result. The only difference is who starts it — sometimes you tell it what to do, sometimes you set a time in advance and it starts on its own.
- You state the goal and acceptance criteria
- The agent reads context and breaks the work down
- It invokes tools, Skills, and Connectors
- It produces and delivers the files
- You review: accept, or send it back
- A schedule or event fires
- The agent reads context and the trigger condition
- It invokes tools, Skills, and Connectors
- It produces files and files or routes them by rule
- It asks for your authorization or notifies you when needed
Both cases run the same work in the middle; only the start and the end differ. A task you hand over comes back for you to accept or reject. A task on a schedule is delivered by the rules you set in advance, and it only comes back to you when it needs authorization or hits an exception.
The picture maps onto the four chapters: Chapter 1 covers the two ways a task starts and the resources it needs; Chapter 2 fixes those practices into a reusable system; Chapters 3 and 4 demonstrate them in real work settings.
The Course Arc
- Meet BasilAOS
- Read the interface and output exits
- Send the first task
- Build the resource map
- Add a Skill and a Work
- Open a Connector
- Connect Apps, Memory, and workspaces
- Advanced systems and authoring
- Office scenarios
- Role scenarios
The first five steps get you through one task on your own: you know what the product is, you can hand work over, and you can direct it to the right resources. The last five extend that reach to external systems, organizational assets, and a full role workflow — and let you write a capability built around your own business.
All Lessons
Chapter 1 · Meet BasilAOS — Your New Work Partner
- 1.1 Meet BasilAOS — It's Not Just Another Chat Box
- 1.2 The Interface: Three Zones and Two Exits
- 1.3 Your First Task: From One Sentence to a Deliverable
- 1.4 Resource Map: The Four Kinds of Resources in Explore
- 1.5 Skill: Add Your First Reusable Capability
- 1.6 Work: Combine Several Skills Into One Work Package
- 1.7 Connector: Reach External Systems and Authorize Access
- 1.8 App
- 1.9 Memory and the Workspace: Two Kinds of Continuity
- 1.10 Choose Resources and Deliverables for Your Task
Chapter 2 · Advanced Systems — Turn One-Time Wins Into a System
- 2.1 Workspaces and Automation: Let the Work Continue
- 2.2 Understanding a Skill: Structure, Progressive Disclosure, and Reuse Value
- 2.3 Creating a Skill: From One Success to a Reusable Method
- 2.4 Tuning Skills: When Standard Isn't Enough
- 2.5 Designing a Work: Assemble Skills Into One Business Chain
- 2.6 Library in Depth: Team Space, Bulk Work, and Knowledge Sources
- 2.7 Base in Depth: Views, Fields, and Sharing Permissions
- 2.8 Appendix: The AI Work System in One Chapter
Chapter 3 · Office Work, Solved — One Task at a Time
- 3.1 Email Management & Structured Follow-ups
- 3.2 Document Summarization & Translation
- 3.3 Data Analysis & Insights
- 3.4 Storytelling — PPT/HTML Generation
- 3.5 Fashion Industry News & Inspiration Search
- 3.6 Personal Folder Organization
Chapter 4 · Role-Specific Scenarios
- 4.1 HR: JD Generation, Resume Screening & Candidate Follow-ups
- 4.2 Legal: Supplier Contract Review, Template Generation & Expiry Tracking
- 4.3 Finance: Invoice Processing, Reconciliation & Reporting
- 4.4 Designers: AI Image Generation for Fashion
- 4.5 E-commerce Operations: Multi-Channel Product Content Generation
- 4.6 Marketing: Competitor Social Media Monitoring, Content Planning & Campaign Execution
- 4.7 Sales: Lead Automation & Multi-Source Prospecting
- 4.8 Merchandise Planning: Product Operations Analysis & Inventory Risk Alerts