Display Name
Learn Agentic Working
Category
Workflows & Knowledge Guides
Sub-Category
General
Primary Link
https://the-good-pixel.github.io/learn-agentic-working/
Author Name
Darren Chiu
Author Link
https://github.com/the-good-pixel
License
MIT
Other License
No response
Description
An open-source playbook (28 chapters) on working with AI agents like Claude Code, Codex, and OpenCode, written for both engineers and non-engineers. Part V splits into role-specific chapters with workflows for PMs, designers, marketers, finance, ops, and engineers. Every workflow is paraphrased from a real working team, with downloadable example files (xlsx, pdf, docx) shipped with the book.
Validate Claims
The book ships with downloadable example files (xlsx, pdf, docx, png) referenced by chapter exercises, so any reader can reproduce a workflow end-to-end. The architecture model introduced in Ch. 2 (You -> Orchestrator -> Model -> Connector -> Real app) is consistent across the entire book. Tool-neutrality is verifiable by spot-checking any chapter: each has an "In other tools" callout citing Codex, OpenCode, Cursor, and Gemini CLI equivalents.
Specific Task(s)
Reader-facing example tasks the book teaches:
- Draft a PRD from messy meeting notes (Ch. 22)
- Reconcile a month of Stripe payouts against an expected schedule (Ch. 23)
- Pull a weekly Google Analytics report and explain the deltas (Ch. 22)
- Run a paid-ads review loop on Google Ads or Meta Ads weekly (Ch. 22)
- Cross-repo bug hunt across a frontend/backend split (Ch. 25)
- Build a product demo video from a running app via Playwright + Remotion (Ch. 20)
Specific Prompt(s)
Each workflow chapter contains paste-able prompt templates. A representative one (Ch. 23, Stripe reconciliation):
"Open the attached Stripe payouts CSV for last month. Cross-check it against the expected payout schedule in our finance sheet. Flag any payouts that are missing, late by more than 2 business days, or have a deviation greater than 1% from the expected amount. Output a short table with: payout date, expected, actual, delta, and a one-line note on what looks off."
Additional Comments
Disclosure: I'm the author. The book is heavily inspired structurally by Learn Harness Engineering (Walking Labs), credited in the README and further-reading appendix. Site source is Markdown under /docs, rendered with VitePress, multilingual-ready.
Recommendation Checklist
Display Name
Learn Agentic Working
Category
Workflows & Knowledge Guides
Sub-Category
General
Primary Link
https://the-good-pixel.github.io/learn-agentic-working/
Author Name
Darren Chiu
Author Link
https://github.com/the-good-pixel
License
MIT
Other License
No response
Description
An open-source playbook (28 chapters) on working with AI agents like Claude Code, Codex, and OpenCode, written for both engineers and non-engineers. Part V splits into role-specific chapters with workflows for PMs, designers, marketers, finance, ops, and engineers. Every workflow is paraphrased from a real working team, with downloadable example files (xlsx, pdf, docx) shipped with the book.
Validate Claims
The book ships with downloadable example files (xlsx, pdf, docx, png) referenced by chapter exercises, so any reader can reproduce a workflow end-to-end. The architecture model introduced in Ch. 2 (You -> Orchestrator -> Model -> Connector -> Real app) is consistent across the entire book. Tool-neutrality is verifiable by spot-checking any chapter: each has an "In other tools" callout citing Codex, OpenCode, Cursor, and Gemini CLI equivalents.
Specific Task(s)
Reader-facing example tasks the book teaches:
Specific Prompt(s)
Each workflow chapter contains paste-able prompt templates. A representative one (Ch. 23, Stripe reconciliation):
"Open the attached Stripe payouts CSV for last month. Cross-check it against the expected payout schedule in our finance sheet. Flag any payouts that are missing, late by more than 2 business days, or have a deviation greater than 1% from the expected amount. Output a short table with: payout date, expected, actual, delta, and a one-line note on what looks off."
Additional Comments
Disclosure: I'm the author. The book is heavily inspired structurally by Learn Harness Engineering (Walking Labs), credited in the README and further-reading appendix. Site source is Markdown under /docs, rendered with VitePress, multilingual-ready.
Recommendation Checklist