01
Clarify work before implementation
Use grilling and shared-language workflows to expose assumptions, define terms, and align on what should be built.
Matt Pocock engineering workflows
CategoryAI & Codex Skills
Matt Pocock Codex Skills
Matt Pocock Codex Skills bring a collection of small, composable engineering workflows into Codex for requirements clarification, shared project language, test-driven development, structured debugging, decomposition, architecture work, issue handling, and concise communication. Download the skills when you want stronger process without handing control to a large framework.
Engineering workflowsRequirementsTDDStructured debugging
01
02
03
04
05
06
07
08
09
02
03
04
05
06
07
08
09
task: complete matt pocock engineering workflows task
inspect:
- requirements and context
- existing standards
- failure and edge cases
verify: outputs + checks + handoffWhy a specialist workflow matters
Fast agent output creates expensive rework when requirements are misaligned, debugging becomes guesswork, tests arrive after the implementation, and one framework tries to own every decision.
Matt Pocock's skill collection is designed as a set of focused workflows that teams can combine and adapt rather than one mandatory process.
This page helps Codex users understand the collection, install the relevant skills, run repository setup, and choose the smallest workflow that fits the current engineering problem.
What the downloadable skills can do
What Matt Pocock Codex Skills can help accomplish
02
Develop with evidence
Apply red-green-refactor loops, reproduction-first debugging, diagnostics, and explicit success checks.
03
Break down complex engineering
Decompose issues, organize parallelizable work, evaluate architecture, and maintain useful project documents.
04
Adapt the collection to a team
Select individual skills, configure issue-tracker and documentation preferences, edit copied files, and preserve team control.
A repeatable working process
How the matt pocock engineering workflows workflow moves from request to verified result
The steps keep context, implementation, and verification visible so the result can be reviewed and repeated.
workflow.statusREADY
Context → Plan → Work → Verify
01
Inspect the available collection
Review current upstream skill names, files, license, setup instructions, dependencies, and changes before installation.
02
Select only relevant workflows
Choose the skills that match requirements, debugging, testing, architecture, issue handling, or communication needs.
03
Run project setup deliberately
Configure issue tracking, labels, document locations, project terminology, and repository-specific expectations.
04
Use and refine in real work
Invoke the focused skill, inspect its results, adapt copied instructions to the team, and update with review.
Useful across real projects
Where Matt Pocock Codex Skills fit
The workflow adjusts to the project, audience, tools, and risk while preserving the same quality standard.
Product and requirements work
Use grilling, terminology, issue triage, and planning workflows before implementation.
Engineering implementation
Apply test-driven development, decomposition, issue execution, and focused verification.
Debugging and diagnosis
Reproduce problems, form hypotheses, inspect evidence, change one cause at a time, and confirm the fix.
Architecture and maintenance
Review design, reduce entropy, document decisions, and improve structure without unnecessary process ownership.
Common requests
Tasks these skills can handle
Start with one defined outcome and provide the source material, constraints, and checks that matter.
01Install selected Matt Pocock skills
02Run repository setup
03Clarify a feature request
04Use red-green-refactor TDD
05Diagnose a difficult bug
06Break down a large issue
07Review architecture
08Adapt a copied skill to team conventions
Who benefits most
Who Is This For?
These skills are designed for people who need dependable matt pocock engineering workflows work with a visible process.
Professional software engineers
Add focused, editable workflows to everyday agent-assisted development.
Engineering teams
Align requirements, terminology, testing, debugging, issues, and documentation without adopting one large framework.
Technical leads
Improve decision quality, decomposition, architecture, and evidence while retaining control of the process.
Codex skill collectors
Install and evaluate individual workflows from a well-known engineering skill repository.
Good to know:
These are Matt Pocock's community-published workflows used with Codex through the open Agent Skills ecosystem. Review the current source, license, scripts, and setup instructions before installing or updating them.
Set up the workflow
Installation Guide
Install the complete skill folder and add the project-specific context before beginning.
Review the current Matt Pocock skills repository
Inspect the README, promoted skills, setup workflow, license, recent changes, and any scripts before installation.
Install through the supported skills CLI
Use the current documented command, select Codex and the specific skills needed, and choose project or global scope deliberately.
Run the repository setup skill
Configure the issue tracker, labels, generated-document location, and other project preferences requested by the collection.
Test selected workflows
Try requirements, TDD, debugging, or decomposition on representative tasks and adapt editable copies where needed.
Before you download
Frequently Asked Questions
Practical answers about capabilities, limits, setup, and review.
Do these skills work with Codex?+
Yes. The open skills installer supports Codex and can copy selected skill folders into a supported project or user scope.
Do I need to install the whole collection?+
No. The collection is designed to be composable, so choose the workflows relevant to your work.
What does the setup skill configure?+
The current repository setup asks about project preferences such as issue tracking, labels, and where generated documents should be stored.
Can I edit the skills?+
Skills copied into your project are editable. Review upstream changes before deciding how to update local adaptations.
Are these official OpenAI skills?+
No. They are community-published engineering skills that can be used with Codex.
Make the work repeatable
Give Codex a composable engineering toolkit for alignment, testing, debugging, decomposition, and architecture without giving up control of the work.
Clear context. Purposeful work. Relevant checks. A result others can understand.