nzt
claudecodexThe generic NZT method, for any work in any domain
8 skills
The NZT method, its agents and its software rules
Plugin nzt-software from marketplace nzt in Ezefeola/nzt-ai. Commands read from the repo's manifest — 2 days ago.
architectureUse when how it is made has to be decided before the work - its parts and boundaries, technologies and practices, the agent's tools, the domain model, one feature's technical design, integrations, decision records, debt, diagrams.
delegateUse before handing work to a subagent - a reading, or a unit the approved plan marked parallel: whether the mode allows it, the prompt, and how to check what comes back.
designUse when the form of what a person perceives or operates has to be decided - layout, content, states and navigation, a mockup or prototype, the design system and voice, a usability review, the end user's manual.
discoveryUse when what it must do and for whom has to be agreed - purpose, rules, scope, vocabulary, specs and stories, changing what was agreed, or reconstructing it from existing work.
learnUse when the user wants to acquire or improve a skill instead of having it built for them: learning, practising, being corrected, assessed, or keeping it from fading.
nztNZT's working method - the rules for every reply and where each request goes. Load it first in every session, before anything else, even for a question, and again after the conversation is compacted.
planUse when a plan has to be built, approved, changed, executed or resumed, when a feature the user accepted is closed, and whenever Plan/state.json is read or written - its schema lives here.
researchUse before stating or building on how an external system, API, library, standard or agent tool (an MCP server) behaves: the version in use, the authoritative source, every claim labelled.
retroUse only when the user asks for a retro of a session: proposals to improve the project and NZT from where the agent stumbled, written to a file outside the repo. Changes nothing else.
shipUse when the work goes to other people - version control, changelog and versions, the release pipeline, releasing, deploying or publishing, monitoring a release, rolling back.
softwareUse for the code of a component Docs/Architecture/tech.md declares - its language, framework and practices (C#, .NET, Blazor, DDD, backend, frontend) - before designing, writing, testing or shipping it.
verifyUse when what was made needs evidence - test plans from a story, test data, running scenarios, exploratory sessions, performance, regression, defects - or reading the work or documents for what does not hold.
explorerRead-only explorer for NZT. Use only when the nzt-delegate skill sends a reading here - surveying unfamiliar code, searching the whole repository, or researching on the web.
workerRuns one approved NZT unit that the plan marked parallel. Use only when the nzt-delegate skill sends a unit here - never for reading, planning or questions.
The generic NZT method, for any work in any domain
8 skills
Step through the last turn's edits or the project's changes against HEAD, one diff at a time
Browse the project like an editor's explorer: folders that open and close, git marks, what Claude reads and writes, and the chosen file beneath
Themes (NZT and fifteen more, dark and light), a usage band with NZT's 30/40 context marks, and the conversation redrawn
nzt-software mod: live pane with the plan and state from Plan/state.json
Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
15 skills · hooks
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
Open-source AI job search that runs locally in your AI coding CLI: scan job portals, evaluate listings into a structured A-H report with a 1-5 score, tailor your CV, track applications
5 skills · hooks
Cross-session memory and token savings for coding agents.
6 skills · 1 agent · 1 MCP server · hooks
Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.
1 MCP server
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.