mattwynne

Design trustworthy, maintainable acceptance-test automation across domain, service, and UI boundaries in any technology stack.

Plan a behaviour-changing iteration through adversarial example mapping, Gherkin, collaborative domain modelling, Matt-approved ADRs, plan assembly, and publication.

Choose domain-only BDD implementation or full iteration delivery for a published ready or validated plan, then ask for explicit launch approval and run the guarded Fabro command.

Start planning a product or engineering iteration, classify it as behaviour-changing or technical/refactoring, and route into the appropriate specialist planning skill.

Plan a technical or refactoring iteration that preserves observable behaviour, with explicit capability, constraints, architecture decisions, proof, and publication.

Investigate delivery-workflow friction or a docs/kaizen note, creating the observation first when no matching note exists; establish evidence-backed causes and options without implementing a fix.

Implement and validate a countermeasure for an investigated docs/kaizen note, then document its resolution. Use when asked to fix or resolve a kaizen note, not for investigation-only requests.

Offer to capture and, when accepted, write a docs/kaizen observation when we notice friction, imperfections, waste, opacity, brittleness, or failure in the pipeline/factory/workflows/tooling used to create the product. Use for workflow/tooling problems, not…

Use before writing Gherkin when exploring a story's behaviour, rules, examples, questions, scope, or readiness with stakeholders

Use when writing or reviewing Gherkin scenarios, especially after discovering examples or edge cases that reveal a business rule

Use when a feature feels under-tested, after implementing new functionality, or before a release to discover edge cases, UX issues, and bugs through hands-on CLI exploration

Use when running yx commands that create, modify, or delete yaks outside of real project work — provides an isolated temp environment

Pick up ready leaf yaks and implement them — dispatch subagents in isolated worktrees, then merge back to main

Prepare a yak for implementation by establishing spec, acceptance criteria, and breaking into sub-yaks

Discover work structure by approaching goals and finding blockers — emergent planning through action, not top-down decomposition

Use when a yak needs requirements, examples, and a plan before implementation - prepares a yak so it's ready for subagent-driven development

Communicate with people in plain, concrete language. Use when explaining work, reporting progress, describing risks, asking questions, or handing work back to a human, especially after a long technical task.

Improve, rewrite, critique, or draft prose with a concise, vigorous, reader-centered style inspired by classic usage guidance including Strunk's public-domain The Elements of Style. Use for essays, emails, docs, blog posts, memos, announcements, and any…

Use when creating or editing Architecture Decision Records - guides thinking through decisions with stakeholders before documenting the outcome

Investigate and resolve docs/kaizen problem observation notes. Use when the user asks to fix a kaizen note, investigate root cause, suggest repair options, or apply an obvious workflow/tooling fix and update the note with the resolution.

Offer to capture and, when accepted, write a docs/kaizen observation when we notice friction, imperfections, waste, opacity, brittleness, or failure in the pipeline/factory/workflows/tooling used to create the product. Use for workflow/tooling problems, not…

Capture delivery-process friction as a kaizen note and immediately run the kaizen-fix workflow on it. Use when Matt asks to JFDI a kaizen issue, wants a kaizen note and fix in one pass, or asks to turn an observed workflow/tooling problem directly into an…

Use when distilling practical design heuristics from code reviews, PR discussions, design sessions, architecture debates, incident reviews, or prior team decisions

Run a parallel multi-model review with Gemini, Codex, and Claude, then synthesize the reports into a single decision-oriented review. Use when the user asks for an ensemble review, three-way review, deep review, or independent second opinions on code, design,…

Ship verified PA for BB repository changes and get feedback from the running BB installation. Use after implementing plugin, marketplace, or documentation changes that are ready for delivery.

Review a bounded plan, design, code change or document using a supplied rubric and up to three independent model families; synthesize evidence, disagreements and actionable questions. Invoke explicitly for an ensemble or multi-model review, not routine…

Plan, implement, or review settings, onboarding, connection, error, or approval UI for PA for BB plugins. Use for any plugin UI change in this repository.

Pick the next ready iteration plan from docs/iterations and execute it through the Fabro iteration-implementation workflow. Use when Matt asks to implement, run, execute, or start the next ready iteration.

Interview Matt about the next product/dev iteration, turn the discussion into a focused iteration plan, and submit it to the project's Fabro plan-validation workflow. Use when planning the next iteration, shaping work before implementation, or preparing a…

Explain how to use BB Hands-Free's iPhone tap-to-talk view and its safety limits.

Find prior AI coding-agent/Pi sessions, recover context, locate artifacts created in earlier sessions, or reopen old sessions using the holocron CLI. Use when Matt mentions holocron, prior sessions, previous agent work, missing Google Docs/files/URLs, or asks…