SeemSeam/agent-roles-spec - Page 2

su-archive

在 review 通过后固化完成证据、风险和后续建议。

Source text: Chinese

su-batch

建立 autonomous-batch 授权范围和停止边界的指令入口。

Source text: Chinese

su-cancel

取消 Requirement 或子任务的大状态指令入口。

Source text: Chinese

su-defer

暂缓 Requirement 或子任务的大状态指令入口。

Source text: Chinese

su-dispatch

把已确认的子任务派给执行 agent 的指令入口。

Source text: Chinese

su-flow

CCB 主流程入口：按用户意图在 7 个节点中选择工作模式，推进需求分析、设计、拆分或后续执行。

Source text: Chinese

su-init

初始化项目级 CCB plugin 工作区，使 plugin 可脱离 Console 独立运行。

Source text: Chinese

su-materialize-requirement

将用户已审查通过的 Requirement breakdown draft 物化为子任务。

Source text: Chinese

su-plan

/ccb:su-flow 的 deprecated alias；保留到 v1.5 grace window。

Source text: Chinese

su-quick-archive

对低风险已通过审查事项执行快速归档，但仍保留证据和审计。

Source text: Chinese

su-reactivate

重新激活 cancelled/deferred 主体的大状态指令入口。

Source text: Chinese

su-reconcile

AI-native 自检 docs 人读真相源、.ccb 协调件与 Console 投影漂移，并按审批结果执行修复。

Source text: Chinese

su-resume

从 docs 人读真相源与 .ccb 协调件恢复 paused/deferred/in-progress 主体的上下文。

Source text: Chinese

su-review

审查执行回执、diff 和验证证据，决定通过、返工、replan 或升级用户。

Source text: Chinese

su-revise-breakdown

根据用户反馈重写 Requirement breakdown draft。

Source text: Chinese

su-status

从 docs 人读真相源与 .ccb 协调件读取当前项目、Requirement 或子任务状态。

Source text: Chinese

mobile-virtual-device-lab

Plan, install, boot, control, inspect, reset, and clean up Android emulators, Android Virtual Devices, iOS Simulators, Expo Orbit, React Native, Expo, and Flutter virtual-device workflows with explicit host approval.

code-review-and-quality

Conducts multi-axis code review. Use before merging any change. Use when reviewing code written by yourself, another agent, or a human. Use when you need to assess code quality across multiple dimensions before it enters the main branch.

improve-codebase-architecture

Find deepening opportunities in a codebase, informed by the domain language in CONTEXT.md and the decisions in docs/adr/. Use when the user wants to improve architecture, find refactoring opportunities, consolidate tightly-coupled modules, or make a codebase…

receiving-code-review

Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation

requesting-code-review

Use when completing tasks, implementing major features, or before merging to verify work meets requirements

role-blueprint

Produce a machine-checkable Agent Role blueprint before writing Role source; use after research, candidate scoring, or source ingestion to define Role id, aliases, catalog level, memory, skills, references, tools, adapters, permissions, provenance, validation…

role-candidate-score

Score and filter candidate sources before turning them into Agent Roles; use after role research when comparing skills, plugins, repositories, tools, prompts, or workflow sources for relevance, authority, license, host fit, security, testability, and…

role-creation-audit

Create, audit, repair, or explain Agent Role source for spec compliance and catalog readiness; use for Role authoring, Role source review, memory/skill/prompt optimization, catalog checks, and routing broad research or external source conversion through the…

role-source-ingest

Analyze external skill, plugin, prompt, tool, or workflow source before converting it into an Agent Role; produce inventory, source classification, provenance, split decision, blueprint gate, and validation plan before writing Role source.

browser-quality

Use to validate a running frontend with browser evidence: visual QA, responsive screenshots, accessibility snapshots, console/network issues, layout collisions, and LCP/INP/CLS performance risks.

ui-ux-pro-max

UI/UX design intelligence for web and mobile. Includes 50+ styles, 161 color palettes, 57 font pairings, 161 product types, 99 UX guidelines, and 25 chart types across 10 stacks (React, Next.js, Vue, Svelte, SwiftUI, React Native, Flutter, Tailwind,…

mobile-app-brief

Create a mobile app implementation brief. Use when starting a mobile app or feature, clarifying iOS and Android platform scope, choosing native versus cross-platform stack, or turning requirements, screenshots, Figma context, or rough ideas into a build plan.

mobile-component-system

Select and apply mobile component systems, design tokens, icons, illustrations, fonts, animation assets, and UI kits. Use for app visual systems, reusable mobile components, theming, asset-library choices, and license or bundle-size tradeoffs.

mobile-device-quality

Validate mobile apps with simulator, emulator, device, accessibility, performance, crash, log, and form-factor evidence. Use for mobile QA, safe-area bugs, keyboard issues, gesture failures, layout clipping, jank, startup time, memory, network, and…

mobile-library-curation

Curate public mobile skills, UI kits, icon sets, animation runtimes, design resources, sample apps, and asset libraries. Use when choosing what to install, vendor, reference, or avoid for mobile app development.

mobile-release-readiness

Review mobile app release readiness for App Store, Google Play, TestFlight, Play testing tracks, EAS, CI/CD, privacy permissions, metadata, screenshots, subscriptions, in-app purchases, account deletion, tracking consent, and store rejection risk.

mobile-stack-patterns

Apply mobile implementation patterns for React Native, Expo, Flutter, SwiftUI, Jetpack Compose, native modules, routing, state, data fetching, storage, offline behavior, and platform-specific code. Use when coding or reviewing mobile app architecture.

mobile-ux-flow

Design native-feeling mobile flows. Use for mobile navigation, tabs, stacks, sheets, gestures, safe areas, keyboard behavior, haptics, permissions, onboarding, empty/error/loading/offline states, and platform-specific iOS or Android interaction polish.

role-setup

Check, plan, apply, or repair role-scoped runtime setup for the current mounted Role. Use when the user asks to activate role tools, enable MCP/plugin/provider configuration, run role_setup, inspect isolation, verify private tool installation, or repair…

agy-frontend-delegate

Use when the user wants Google Antigravity CLI (`agy`) to attempt or compare a bounded frontend implementation in an isolated worktree, with diff review before merge.

component-composition

Use before building UI with local components, Storybook, shadcn, Radix, a private registry, or another component library; verifies available components, props, states, and composition choices.

demo-kb-curation

Use to curate link-only frontend demos, inspiration sources, pattern references, and component examples with source authority, license posture, tags, and do-not-copy boundaries.

design-system-tokens

Use for design tokens, theming, light/dark or multi-brand support, token drift, component-system alignment, Figma variables, or Style Dictionary/DTCG token workflows.

figma-to-code

Use when Figma context, selected frames, variables, components, or screenshots should be mapped to local frontend implementation without copying private Figma state into Role source.