Local-first agent platform: board + role agents + agent CLI runs (Claude Code, Cursor, Antigravity, OpenCode) or local and API models (Ollama, LM Studio), all on your own Mac
This repository also configures its own agents. See what tasktrooper tells them →
Skill Claude CodeCodex
Use when building Flutter screens - compose small stateless widgets, keep layout declarative, decide layout by available width not device/orientation, separate presentation from business logic, and use the theme not hardcoded styles.
Skill Claude CodeCodex
Use when building or changing any interactive mobile UI - VoiceOver/TalkBack labels, grouping, headings, focus order, Dynamic Type/font scale, reduce motion, contrast, and the automated check for each stack.
Skill Claude CodeCodex
Use when the mobile repository has no theme tokens or component library yet (a new app, an "Apply design system" task, or a task that asks for one), before building any screen - writes the approved design system's generated files from getdesignsystem (files true) and builds the theme from them when there is one.
Skill Claude CodeCodex
Use when adding a screen, a tab, a deep link, or back-navigation handling - the per-stack navigation APIs, state preservation, auth-boundary resets, and how to verify a deep link actually opens the right screen.
Skill Claude CodeCodex
Use when designing or building any mobile screen or component a user sees (Flutter, SwiftUI or Compose) - platform conventions, hierarchy, type that scales, colour roles and dark mode, safe areas and insets, forms and keyboard, states, feedback, and the generic-mobile anti-patterns.
Skill Claude CodeCodex
Use when you are about to hand off any change to a screen or component on a mobile repository - the size, text-scale and dark-mode matrix, the exact tool calls to see your own build, and the checklist that decides whether the screen is done.
Skill Claude CodeCodex
Use when starting a mobile task that could be Flutter or native - decide from the existing app, platform-specific needs, and the analiz task.
Skill Claude CodeCodex
Use when a screen must work without network or a mutation must survive a lost connection - local storage choice, the outbox pattern with idempotency keys, background replay, and how to test it with a fake clock and fake network.
Skill Claude CodeCodex
Use when a feature needs camera, location, photos, notifications, contacts, microphone, Bluetooth, or local network - the per-platform request flow, every denial state, and how to actually test the denied path when the shared test device auto-grants permissions.
Skill Claude CodeCodex
Use when building native iOS with SwiftUI - small composable views, observable state models, atomic components, Swift 6.2+ concurrency, and system styling over hardcoded values.
Skill Claude CodeCodex
Use when writing acceptance criteria for a task - express each as an observable Given/When/Then that QA can execute, including negative cases.
Skill Claude CodeCodex
Use when deciding whether a request needs an analiz task before implementation - the conditions that require the architect's analysis versus going straight to implementation.
Skill Claude CodeCodex
Use when opening an analiz task for the system-architect - the required fields, what the criteria may say, and what the description must hand over.
Skill Claude CodeCodex
Use when the stakeholder asks a factual or status question about the workspace - call the matching read tool and answer conversationally, without creating tasks or asking for approval.
Skill Claude CodeCodex
Use when ordering board tasks - set priority/blockedby/deploydependson by severity and dependency, split MoSCoW within a feature, and use RICE/ICE only for a genuine "what next" call across several features.
Skill Claude CodeCodex
Use when moving tasks or reporting status - the kanban column semantics, which columns are human/architect/QA gates, and where the PM actually acts.
Skill Claude CodeCodex
Use when you create an implementation board task directly - the required fields, one-role-one-deliverable rule, and dependency ordering.
Skill Claude CodeCodex
Use when planning a delivery request as the only orchestration agent - one board-writing subtask, backlog first, one approval question, then move approved tasks to todo.
Skill Claude CodeCodex
Use when a task is in pmuat - verify every acceptance criterion and the human's comments yourself in the browser, on a device, or over a loopback HTTP request, record a verdict per criterion, then move to humanuat or needrevision.
Skill Claude CodeCodex
Use when the stakeholder asks about an analysis outcome or its implementation tasks - understand the human approval gate and review the tasks the architect created once the human approved.
Skill Claude CodeCodex
Use when organizing the workspace - create/rename initiative projects and link repositories to them directly, since these are factual actions needing no analiz or approval.
Skill Claude CodeCodex
Use when the stakeholder asks for release notes or a changelog - group done/released work into Added, Changed, Deprecated, Removed, Fixed and Security, in outcome language, with task references.
Skill Claude CodeCodex
Use when a product decision or feature needs a written record - what to capture in a task document (PRD, decision record) and where to attach it, since TaskTrooper has no epic.
Skill Claude CodeCodex
Use when capturing what a feature must do - write testable user-story requirements that state WHAT and WHY, never HOW, with assumptions made explicit.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: