TeamBrain

Non-technical install guide and onboarding walkthrough for TeamBrain. Reads INSTALL.md and narrates each step in plain Chinese for non-developers. Handles install errors by matching against known error patterns and explaining the fix aloud. Use when: "install", "onboarding", "walkthrough", "pnpm 是什么", "怎么安装", "我装不上", "non-technical install", "帮我安装", "安装步骤", "第一次使用", "how do I install", "set up this project".

[ARCHIVED] The visual-proof-pr local dev-loop skill has been archived to docs/legacy/skills/visual-proof-pr/. This stub stays at the original path so references to .claude/skills/visual-proof-pr/ do not break. Do NOT trigger this skill — it no longer carries an active workflow.

Visual-proof guided PR workflow: design-shotgun variants → design-html finalization → Chrome pop-open human verify → implement → /review → squash-merge PR. Use when: "implement visual-proof PR", "visual proof workflow", "design → PR" Triggers: visual proof PR, design to PR, implement with visual proof

TeamBrain repo **24h window delta** report — 拉过去 N 小时窗口内 PR/issue 的创建/合并/关闭事件，按主题（features/bugfixes/docs/infra）分组，鸭语汇报。仅当用户问 `what happened since yesterday` / `explore today's issues and PRs` / `daily standup` / `/daily-standup` / `今天的 issue 和 PR` / `昨天到今天发生了啥` / `show me yesterday's activity` / `过去 24 小时 GitHub 活动` / `9 点到 9 点发生了什么` 这类**时间窗口 delta 类**问题时触发。**不要在以下情况触发**：(a) 用户问当前快照（`status of all issues` / `所有 open issue` → 用 `repo-issues-status` skill）；(b) 用户问今日开工晨报（`/newday` → 用 `newday` skill，覆盖 worktree + 当前开放工作而非 24h delta）；(c) 用户问单 issue/PR 状态（`#330 怎么样` → 用 `gh issue view`/`gh pr view`）。匹配命中后默认拉 yesterday 09:00 +0800 → today 09:00 +0800 的窗口，事件按 PRs merged / created / closed-without-merge / 还开着 + Issues created / closed / 还开着 六桶分组，再按主题二次归类，全文用 呷呷~ 鸭鸭 口吻包裹。

TeamBrain repo **24h window delta** report — 拉过去 N 小时窗口内 PR/issue 的创建/合并/关闭事件，按主题（features/bugfixes/docs/infra）分组，鸭语汇报。仅当用户问 `what happened since yesterday` / `explore today's issues and PRs` / `daily standup` / `/daily-standup` / `今天的 issue 和 PR` / `昨天到今天发生了啥` / `show me yesterday's activity` / `过去 24 小时 GitHub 活动` / `9 点到 9 点发生了什么` 这类**时间窗口 delta 类**问题时触发。**不要在以下情况触发**：(a) 用户问当前快照（`status of all issues` / `所有 open issue` → 用 `repo-issues-status` skill）；(b) 用户问今日开工晨报（`/newday` → 用 `newday` skill，覆盖 worktree + 当前开放工作而非 24h delta）；(c) 用户问单 issue/PR 状态（`#330 怎么样` → 用 `gh issue view`/`gh pr view`）。匹配命中后默认拉 yesterday 09:00 +0800 → today 09:00 +0800 的窗口，事件按 PRs merged / created / closed-without-merge / 还开着 + Issues created / closed / 还开着 六桶分组，再按主题二次归类，全文用 呷呷~ 鸭鸭 口吻包裹。

CEO 鸭的"今日开工"晨报 — 从远程拉取代码、检查所有 worktree、汇总开放 PR 与 issue，并用非技术语言告诉鸭老板"今天有什么决策要做、有什么活儿可以推进"。手动触发：/newday。

Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me".

Pop a clickable ChatGPT and Claude.ai URL that prefills a grill-me prompt referencing a public GitHub issue, so a human can finish the grill in a browser tab. Use when the user wants to grill an issue in a web LLM (no API, no local Claude needed). Trigger phrases include "grill via web", "grill issue N in chatgpt", "/grill-via-web", "give me a chatgpt url to grill <issue>", or "pop a grill URL for issue N".

Grilling session that challenges your plan against the existing domain model, sharpens terminology, and updates documentation (CONTEXT.md, ADRs) inline as decisions crystallise. Use when user wants to stress-test a plan against their project's language and documented decisions.

TeamBrain FIXEDFLOW step-3-to-5 driver. Reads a grill-ready issue, creates a worktree, implements per the grill comment, loops /review until PASS, opens a normal PR, squash-merges, cleans up. Invoked manually by a maintainer in a Claude Code session — there is no watcher, no background dispatcher, no automatic trigger. Do NOT invoke unless an issue has been verified to have a valid grill comment + grill-ready label.

TeamBrain 仓库**全量 issue 全景**报告（all-scope only）。仅当用户问"status of all issues" / "show me all issues" / "what issues are open" / "all open issues" / "issue 状态（全景）" / "所有 issue 状态" / "issue 全景" / "仓库 issue 概览" / "还有哪些 open issue" 这类**仓库级、多 issue、概览类**问题时触发。**不要在以下情况触发**：(a) 用户问单个 issue 的状态（"what's the status of issue

Remote bootstrap of a new colleague's Mac via SSH. Installs Homebrew, Node, and `@anthropic-ai/claude-code`, then completes interactive `claude login` over a tmux session whose OAuth URL the human relays to the colleague. Manually invoked via /onboard only, so it never auto-triggers from chat. Use when a teammate needs Claude Code stood up on their Mac and the operator has Tailscale SSH access plus a sudo password the teammate shared out-of-band.

By-hand trigger only. When the user types /duck, explain the current topic / latest answer to a cute CEO non-coder duck. Do not auto-trigger.

By-hand only — DO NOT auto-trigger. When the user types /reverification (or asks for "reverify this", "independent verification", "LLM-uncheatable check", "第三方再验证", "用 repo 之外的方法验证"), re-verify recently-completed work using verification methods that **existed before this repo existed** and that an LLM agent cannot fabricate. Pick 3-5 verifications from a 10-family menu (hardware / cryptographic / blockchain / external real-system / mathematical / temporal / content authenticity / physical / network / third-party trust), dispatch them via real tools that capture raw evidence, and report PASS / FAIL / INCONCLUSIVE based ONLY on captured evidence — never on internal reasoning. Refuse to fall back on TeamBrain's bespoke verifiers (judge harness, fastprobe, /review, feature-verification gates) because those did NOT exist before the repo.

TeamBrain claim-an-issue → merged-code routing。给「我要 claim 一个 issue / 想把 issue 推到合并 / 怎么走 FIXEDFLOW」这类 prompt 一个统一入口表，链回 docs/FIXEDFLOW.md 等 canonical doc。Use when user says "claim issue", "走 FIXEDFLOW", "how do I get my issue merged", "issue 怎么自动变成 PR", "claim an issue".

Remote bootstrap of a new colleague's Mac via SSH — installs Homebrew, Node, and `@anthropic-ai/claude-code`, then completes interactive `claude login` over a tmux session whose OAuth URL the human relays to the colleague. Manually invoked via /onboard only (frontmatter pins `disable-model-invocation: true`), so it never auto-triggers from chat. Use when a teammate needs Claude Code stood up on their Mac and the operator has Tailscale SSH access plus a sudo password the teammate shared out-of-band.

Design finalization: generates production-quality Pretext-native HTML/CSS. Works with approved mockups from /design-shotgun, CEO plans from /plan-ceo-review, design review context from /plan-design-review, or from scratch with a user description. Text actually reflows, heights are computed, layouts are dynamic. 30KB overhead, zero deps. Smart API routing: picks the right Pretext patterns for each design type. Use when: "finalize this design", "turn this into HTML", "build me a page", "implement this design", or after any planning skill. Proactively suggest when user has approved a design or has a plan ready. (gstack) Voice triggers (speech-to-text aliases): "build the design", "code the mockup", "make it real".

Design shotgun: generate multiple AI design variants, open a comparison board, collect structured feedback, and iterate. Standalone design exploration you can run anytime. Use when: "explore designs", "show me options", "design variants", "visual brainstorm", or "I don't like how this looks". Proactively suggest when the user describes a UI feature but hasn't seen what it could look like. (gstack)

HyperFrames CLI tool — hyperframes init, lint, inspect, preview, render, transcribe, tts, doctor, browser, info, upgrade, compositions, docs, benchmark. Use when scaffolding a project, linting, validating, inspecting visual layout in compositions, previewing in the studio, rendering to video, transcribing audio, generating TTS, or troubleshooting the HyperFrames environment.

Asset preprocessing for HyperFrames compositions — text-to-speech narration (Kokoro), audio/video transcription (Whisper), and background removal for transparent overlays (u2net). Use when generating voiceover from text, transcribing speech for captions, removing the background from a video or image to use as a transparent overlay, choosing a TTS voice or whisper model, or chaining these (TTS → transcribe → captions). Each command downloads its own model on first run.

Create video compositions, animations, title cards, overlays, captions, voiceovers, audio-reactive visuals, and scene transitions in HyperFrames HTML. Use when asked to build any HTML-based video content, add captions or subtitles synced to audio, generate text-to-speech narration, create audio-reactive animation (beat sync, glow, pulse driven by music), add animated text highlighting (marker sweeps, hand-drawn circles, burst lines, scribble, sketchout), or add transitions between scenes (crossfades, wipes, reveals, shader transitions). Covers composition authoring, timing, media, and the full video production workflow. For CLI commands (init, lint, preview, render, transcribe, tts) see the hyperframes-cli skill.

Use mmx to generate text, images, video, speech, and music via the MiniMax AI platform. Use when the user wants to create media content, chat with MiniMax models, perform web search, or manage MiniMax API resources from the terminal.

Translate an existing Remotion (React-based) video composition into a HyperFrames HTML composition. Use ONLY when the user explicitly asks to port, convert, migrate, translate, or rewrite a Remotion composition as HyperFrames — for example "port my Remotion project to HyperFrames", "convert this Remotion code to HyperFrames", "migrate from Remotion", "translate this Remotion comp", or "rewrite this as HyperFrames HTML". Do NOT use when (a) the user is authoring a NEW HyperFrames composition, even if they have or are A/B-testing a similar Remotion video; (b) the user mentions Remotion in passing without asking for migration; (c) the user shares Remotion code as reference material rather than asking for a translation; (d) the user asks for "the same video as my Remotion one" without explicitly asking to migrate the source — treat that as a fresh HyperFrames build. When in doubt, default to authoring a native HyperFrames composition with the `hyperframes` skill instead. Skill detects unsupported patterns (useSta

Turn the current conversation context into a PRD and publish it to the project issue tracker. Use when user wants to create a PRD from the current context.

Post-deploy canary monitoring. Watches the live app for console errors, performance regressions, and page failures using the browse daemon. Takes periodic screenshots, compares against pre-deploy baselines, and alerts on anomalies. Use when: "monitor deploy", "canary", "post-deploy check", "watch production", "verify deploy". (gstack)

YC Office Hours — two modes. Startup mode: six forcing questions that expose demand reality, status quo, desperate specificity, narrowest wedge, observation, and future-fit. Builder mode: design thinking brainstorming for side projects, hackathons, learning, and open source. Saves a design doc. Use when asked to "brainstorm this", "I have an idea", "help me think through this", "office hours", or "is this worth building". Proactively invoke this skill (do NOT answer directly) when the user describes a new product idea, asks whether something is worth building, wants to think through design decisions for something that doesn't exist yet, or is exploring a concept before any code is written. Use before /plan-ceo-review or /plan-eng-review. (gstack)

CEO/founder-mode plan review. Rethink the problem, find the 10-star product, challenge premises, expand scope when it creates a better product. Four modes: SCOPE EXPANSION (dream big), SELECTIVE EXPANSION (hold scope + cherry-pick expansions), HOLD SCOPE (maximum rigor), SCOPE REDUCTION (strip to essentials). Use when asked to "think bigger", "expand scope", "strategy review", "rethink this", or "is this ambitious enough". Proactively suggest when the user is questioning scope or ambition of a plan, or when the plan feels like it could be thinking bigger. (gstack)

Post-deploy canary monitoring. Watches the live app for console errors, performance regressions, and page failures using the browse daemon. Takes periodic screenshots, compares against pre-deploy baselines, and alerts on anomalies. Use when: "monitor deploy", "canary", "post-deploy check", "watch production", "verify deploy". (gstack)

Design finalization: generates production-quality Pretext-native HTML/CSS. Works with approved mockups from /design-shotgun, CEO plans from /plan-ceo-review, design review context from /plan-design-review, or from scratch with a user description. Text actually reflows, heights are computed, layouts are dynamic. 30KB overhead, zero deps. Smart API routing: picks the right Pretext patterns for each design type. Use when: "finalize this design", "turn this into HTML", "build me a page", "implement this design", or after any planning skill. Proactively suggest when user has approved a design or has a plan ready. (gstack) Voice triggers (speech-to-text aliases): "build the design", "code the mockup", "make it real".

Design shotgun: generate multiple AI design variants, open a comparison board, collect structured feedback, and iterate. Standalone design exploration you can run anytime. Use when: "explore designs", "show me options", "design variants", "visual brainstorm", or "I don't like how this looks". Proactively suggest when the user describes a UI feature but hasn't seen what it could look like. (gstack)

YC Office Hours — two modes. Startup mode: six forcing questions that expose demand reality, status quo, desperate specificity, narrowest wedge, observation, and future-fit. Builder mode: design thinking brainstorming for side projects, hackathons, learning, and open source. Saves a design doc. Use when asked to "brainstorm this", "I have an idea", "help me think through this", "office hours", or "is this worth building". Proactively invoke this skill (do NOT answer directly) when the user describes a new product idea, asks whether something is worth building, wants to think through design decisions for something that doesn't exist yet, or is exploring a concept before any code is written. Use before /plan-ceo-review or /plan-eng-review. (gstack)

CEO/founder-mode plan review. Rethink the problem, find the 10-star product, challenge premises, expand scope when it creates a better product. Four modes: SCOPE EXPANSION (dream big), SELECTIVE EXPANSION (hold scope + cherry-pick expansions), HOLD SCOPE (maximum rigor), SCOPE REDUCTION (strip to essentials). Use when asked to "think bigger", "expand scope", "strategy review", "rethink this", or "is this ambitious enough". Proactively suggest when the user is questioning scope or ambition of a plan, or when the plan feels like it could be thinking bigger. (gstack)