Agentic Coding Starter Manual
A beginner-friendly manual for using AI coding agents without losing the project, the goal, or your confidence. Start with the manual, then use the Coding Lesson of the Day series for small practical exercises.
Start Here
Coding Lesson of the Day: Project State
A 5-minute beginner exercise: create one file that helps an AI agent understand your project before it edits anything.
Learning outcome: create a tiny project-memory loop with `PROJECT_STATE.md`.
Try the lesson
Field Notes From a First-Time Agentic Builder
Where new agentic coders get tripped up, what questions they ask, and how to turn confusion into a repeatable build loop.
Learning outcome: diagnose whether you need context, a decision, a test, or a smaller next task.
Read the field notes
Building AI Tools With AI to Advocate for Humanity
The launch thesis: Basic and Pro should help people build useful software while protecting agency, ownership, and judgment.
Learning outcome: understand the mission and why local-first project memory matters.
Read the launch note
Own more of the stack
Path to Freedom
The practical resource hub for local-first project memory, beginner agentic coding, open-weight AI literacy, privacy habits, and human-first tools.
Learning outcome: choose one concrete step that increases ownership without turning freedom into a vague slogan.
Open the resource hub
Local and Open AI
Learn how open-weight model literacy, safe community examples, evals, and the Trinity Model Scorecard fit the DigitalQuill movement.
Learning outcome: judge local/open AI tools by Outcome, Agency, and AI Hygiene before trusting them.
Explore local AI
Leaving Windows
A practical ladder for thinking about operating systems, tool dependence, and when switching is worth the effort.
Learning outcome: separate useful independence from expensive distraction.
Read the guide
Full Archive
What Is an AI Agent?
A plain-language explanation of what agents can do, what they still need from a human, and how to avoid treating autonomy like magic.
Learning outcome: set realistic expectations before starting a build.
Read the explainer
The AI Popsicle Index
A simple value check for agentic work: did the AI help a real person ship, decide, learn, or recover time?
Learning outcome: judge AI output by usefulness instead of volume.
Read the concept
Value-Based Spending for AI Tools
How to think about AI costs, tool subscriptions, and whether a paid workflow is actually producing value.
Learning outcome: decide when spending on AI tools is justified.
Read the guide
The Efficiency Trap
A reminder that efficiency is not the whole point. AI tools should help people live, learn, decide, and build with more agency.
Learning outcome: separate meaningful progress from pure optimization.
Read the reflection
Built With Its Own Workflow
How the product uses its own files, dashboard checks, and launch learning loop to keep the build honest.
Learning outcome: see how the studio workflow applies to a real product build.
Read the build note
If You Are New, Read in This Order
1. Understand the toolStart with "What Is an AI Agent?" so the promise and limits are clear.
2. Learn the workflowRead the free manual, then try the Coding Lesson of the Day: one goal, one next task, one check.
3. Judge valueUse the Popsicle Index and value-based spending notes to keep the project grounded.
4. Own the pathUse Path to Freedom resources to pick one practical ownership step: local AI, privacy habits, open files, or safer tools.