yikart/AiToEarn

Respond terse like smart caveman. All technical substance stay. Only fluff die.

## Persistence

ACTIVE EVERY RESPONSE once triggered. No revert after many turns. No filler drift. Still active if unsure. Off only when user says "stop caveman" or "normal mode".

## Rules

Drop: articles (a/an/the), filler (just/really/basically/actually/simply), pleasantries (sure/certainly/of course/happy to), hedging. Fragments OK. Short synonyms (big not extensive, fix not "implement a solution for"). Abbreviate common terms (DB/auth/config/req/res/fn/impl). Strip conjunctions. Use arrows for causality (X -> Y). One word when one word enough.

Technical terms stay exact. Code blocks unchanged. Errors quoted exact.

Pattern: `[thing] [action] [reason]. [next step].`

# Diagnose

A discipline for hard bugs. Skip phases only when explicitly justified.

When exploring the codebase, use the project's domain glossary to get a clear mental model of the relevant modules, and check ADRs in the area you're touching.

## Phase 1 — Build a feedback loop

**This is the skill.** Everything else is mechanical. If you have a fast, deterministic, agent-runnable pass/fail signal for the bug, you will find the cause — bisection, hypothesis-testing, and instrumentation all just consume that signal. If you don't have one, no amount of staring at code will save you.

Spend disproportionate effort here. **Be aggressive. Be creative. Refuse to give up.**

### Ways to construct one — try them in roughly this order

Interview me relentlessly about every aspect of this plan until we reach a shared understanding. Walk down each branch of the design tree, resolving dependencies between decisions one-by-one. For each question, provide your recommended answer.

Ask the questions one at a time.

If a question can be answered by exploring the codebase, explore the codebase instead.

<what-to-do>

Interview me relentlessly about every aspect of this plan until we reach a shared understanding. Walk down each branch of the design tree, resolving dependencies between decisions one-by-one. For each question, provide your recommended answer.

Ask the questions one at a time, waiting for feedback on each question before continuing.

If a question can be answered by exploring the codebase, explore the codebase instead.

</what-to-do>

<supporting-info>

## Domain awareness

Write a handoff document summarising the current conversation so a fresh agent can continue the work. Save it to a path produced by `mktemp -t handoff-XXXXXX.md` (read the file before you write to it).

Suggest the skills to be used, if any, by the next session.

Do not duplicate content already captured in other artifacts (PRDs, plans, ADRs, issues, commits, diffs). Reference them by path or URL instead.

If the user passed arguments, treat them as a description of what the next session will focus on and tailor the doc accordingly.

# Improve Codebase Architecture

Surface architectural friction and propose **deepening opportunities** — refactors that turn shallow modules into deep ones. The aim is testability and AI-navigability.

## Glossary

Use these terms exactly in every suggestion. Consistent language is the point — don't drift into "component," "service," "API," or "boundary." Full definitions in [LANGUAGE.md](LANGUAGE.md).

- **Module** — anything with an interface and an implementation (function, class, package, slice).
- **Interface** — everything a caller must know to use the module: types, invariants, error modes, ordering, config. Not just the type signature.
- **Implementation** — the code inside.
- **Depth** — leverage at the interface: a lot of behaviour behind a small interface. **Deep** = high leverage. **Shallow** = interface nearly as complex as the implementation.
- **Seam** — where an interface lives; a place behaviour can be altered without editing in place. (Use this, not "boundary.")
- **Adapter** — a concrete thing satisfying an interface at a seam.

# Prototype

A prototype is **throwaway code that answers a question**. The question decides the shape.

## Pick a branch

Identify which question is being answered — from the user's prompt, the surrounding code, or by asking if the user is around:

- **"Does this logic / state model feel right?"** → [LOGIC.md](LOGIC.md). Build a tiny interactive terminal app that pushes the state machine through cases that are hard to reason about on paper.
- **"What should this look like?"** → [UI.md](UI.md). Generate several radically different UI variations on a single route, switchable via a URL search param and a floating bottom bar.

The two branches produce very different artifacts — getting this wrong wastes the whole prototype. If the question is genuinely ambiguous and the user isn't reachable, default to whichever branch better matches the surrounding code (a backend module → logic; a page or component → UI) and state the assumption at the top of the prototype.

## Rules that apply to both

# Setup Matt Pocock's Skills

Scaffold the per-repo configuration that the engineering skills assume:

- **Issue tracker** — where issues live (GitHub by default; local markdown is also supported out of the box)
- **Triage labels** — the strings used for the five canonical triage roles
- **Domain docs** — where `CONTEXT.md` and ADRs live, and the consumer rules for reading them

This is a prompt-driven skill, not a deterministic script. Explore, present what you found, confirm with the user, then write.

## Process

### 1. Explore

# Test-Driven Development

## Philosophy

**Core principle**: Tests should verify behavior through public interfaces, not implementation details. Code can change entirely; tests shouldn't.

**Good tests** are integration-style: they exercise real code paths through public APIs. They describe _what_ the system does, not _how_ it does it. A good test reads like a specification - "user can checkout with valid cart" tells you exactly what capability exists. These tests survive refactors because they don't care about internal structure.

**Bad tests** are coupled to implementation. They mock internal collaborators, test private methods, or verify through external means (like querying a database directly instead of using the interface). The warning sign: your test breaks when you refactor, but behavior hasn't changed. If you rename an internal function and tests fail, those tests were testing implementation, not behavior.

See [tests.md](tests.md) for examples and [mocking.md](mocking.md) for mocking guidelines.

## Anti-Pattern: Horizontal Slices

# To Issues

Break a plan into independently-grabbable issues using vertical slices (tracer bullets).

The issue tracker and triage label vocabulary should have been provided to you — run `/setup-matt-pocock-skills` if not.

## Process

### 1. Gather context

Work from whatever is already in the conversation context. If the user passes an issue reference (issue number, URL, or path) as an argument, fetch it from the issue tracker and read its full body and comments.

### 2. Explore the codebase (optional)

This skill takes the current conversation context and codebase understanding and produces a PRD. Do NOT interview the user — just synthesize what you already know.

The issue tracker and triage label vocabulary should have been provided to you — run `/setup-matt-pocock-skills` if not.

## Process

1. Explore the repo to understand the current state of the codebase, if you haven't already. Use the project's domain glossary vocabulary throughout the PRD, and respect any ADRs in the area you're touching.

2. Sketch out the major modules you will need to build or modify to complete the implementation. Actively look for opportunities to extract deep modules that can be tested in isolation.

A deep module (as opposed to a shallow module) is one which encapsulates a lot of functionality in a simple, testable interface which rarely changes.

Check with the user that these modules match their expectations. Check with the user which modules they want tests written for.

# Triage

Move issues on the project issue tracker through a small state machine of triage roles.

Every comment or issue posted to the issue tracker during triage **must** start with this disclaimer:

```
> *This was generated by AI during triage.*
```

## Reference docs

- [AGENT-BRIEF.md](AGENT-BRIEF.md) — how to write durable agent briefs
- [OUT-OF-SCOPE.md](OUT-OF-SCOPE.md) — how the `.out-of-scope/` knowledge base works

# Writing Skills

## Process

1. **Gather requirements** - ask user about:
   - What task/domain does the skill cover?
   - What specific use cases should it handle?
   - Does it need executable scripts or just instructions?
   - Any reference materials to include?

2. **Draft the skill** - create:
   - SKILL.md with concise instructions
   - Additional reference files if content exceeds 500 lines
   - Utility scripts if deterministic operations needed

I don't know this area of code well. Go up a layer of abstraction. Give me a map of all the relevant modules and callers, using the project's domain glossary vocabulary.

# Video Analysis

Analyzes video content for understanding and extracts highlight clips using AI multimodal analysis.

## When to Use

Use this skill when:

- Analyzing video content for summary
- Extracting highlight/exciting moments
- Generating video keywords and tags
- Creating trailers from longer videos
- Understanding plot and storyline
- Finding viral-worthy moments

# Video Composition

Combines multiple videos/images into a single video with optional background audio using AI-powered processing.

## When to Use

Use this skill when:

- Merging multiple video clips into one
- Creating a slideshow from multiple images
- Adding background music to a video
- Combining videos with audio tracks
- Stitching together video segments

# Social Media Crawling

Downloads and extracts content from social media platforms.

## Supported Platforms and URL Patterns

| Platform    | URL Pattern Examples                                 |
| ----------- | ---------------------------------------------------- |
| Bilibili    | `bilibili.com/video/BVxxxx`, `b23.tv/xxxxx`          |
| YouTube     | `youtube.com/watch?v=xxx`, `youtu.be/xxx`            |
| TikTok      | `tiktok.com/@user/video/xxx`, `vm.tiktok.com/xxx`    |
| Douyin      | `douyin.com/video/xxx`, `v.douyin.com/xxx`           |
| Xiaohongshu | `xiaohongshu.com/explore/xxx`, `xhslink.com/xxx`     |
| Kuaishou    | `kuaishou.com/short-video/xxx`, `v.kuaishou.com/xxx` |

# Image Editing with Sharp

Performs image editing using the Sharp library for high-performance image processing.

## When to Use

Use this skill when:

- Compositing images (adding QR codes, logos, watermarks to posters)
- Resizing or cropping images
- Rotating or flipping images
- Adjusting brightness, contrast, saturation
- Applying blur or sharpening effects
- Getting image metadata (dimensions, format)

# Video Editing with Track Structure

Performs video editing using the Volcengine cloud editing API with native Track structure.

## ⚠<U+FE0F> CRITICAL: ALL PosX/PosY Are Top-Left Coordinates

**ALL PosX/PosY values in this API are TOP-LEFT corner coordinates, NOT center point!**

This applies to ALL filters: transform, crop, delogo, etc.

**For full-screen video display, you MUST use:**
```json
{ "Type": "transform", "PosX": 0, "PosY": 0, "Width": <canvas_width>, "Height": <canvas_height> }
```

# Thumbnail Extraction

Extracts thumbnails from video URLs for use as cover images.

## When to Use

Use this skill when:

- Publishing video content that requires a cover image
- User doesn't provide a cover/thumbnail for video content
- Before publishing to platforms that require cover images (Kwai, Bilibili)

## Platform Cover Requirements

# Drama Recap Generation

Generates narrated recap videos from short drama clips using AI-powered script generation and voice synthesis.

## When to Use

Use this skill when:

- Creating narrated recaps of short dramas
- Generating drama summary videos with voice-over
- Adding AI-generated commentary to drama clips
- Producing drama explanation/recap content

## Key Features

# Image Generation

Generates images using Gemini AI model (default: gemini-3.1-flash-image-preview, compatible: gemini-3-pro-image-preview).

## Parameters

### Required

- `prompt`: Creative description of the image

### Optional

- `imageUrls`: Reference image URLs for editing
- `imageSize`: Output resolution - "1K", "2K", "4K"

# Video Generation

Generates videos using **Grok**.

## Model Selection Strategy

Use `grok-imagine-video` for video generation.

| Scenario | Use Model | Reason |
|----------|-----------|--------|
| Video 1-15s | `grok-imagine-video` | Grok supports direct 1-15s generation |
| Text-to-video | `grok-imagine-video` | Prompt-only generation |
| Image-to-video | `grok-imagine-video` | Uses one reference image |
| Video >15s | Generate multiple clips, then load `editing-videos` to concatenate | Grok max duration is 15s |

# Content Management

Manages media files, drafts, and content organization.

## When to Use

Use this skill when:

- Saving generated media to library
- Creating drafts for later publishing
- Organizing content into groups
- Managing media files

## Workflow

# Subtitle Removal

Removes hardcoded/burned-in subtitles from videos using OCR detection and AI inpainting.

## When to Use

Use this skill when:

- Removing hardcoded subtitles from videos
- Cleaning up videos with burned-in text
- Preparing videos for re-subtitling
- Removing unwanted text overlays
- Cleaning videos for repurposing

# Video Style Transfer

Converts live-action videos into artistic styles (comic, anime, 3D cartoon) using AI-powered style transfer.

## When to Use

Use this skill when:

- Converting videos to comic/manga style
- Creating anime-style versions of videos
- Applying 3D cartoon effects to videos
- Transforming real-world footage into artistic styles

## Available Styles

# Video Translation

Translates video content to another language using AI voice cloning, lip sync, and subtitle translation.

## When to Use

Use this skill when:

- Translating video content to another language
- Creating dubbed versions of videos
- Adding translated voiceover
- Localizing video for different markets
- Converting Chinese videos to English (or vice versa)

# HARD OUTPUT RULE — READ FIRST

**Generate one separate horizontal image PER section. Always. No exceptions.**

- 1 section requested -> 1 image
- 4 sections requested -> 4 images
- 8 sections requested -> 8 images
- 12 sections requested -> 12 images
- "landing page" with no count -> default to 6 sections -> 6 images
- "full website template" -> default to 8 sections -> 8 images

Each image is one section, generated as its own image call. Never combine multiple sections into one frame. Never return a single tall image that contains the whole page.

If you can only render one image at a time, output them sequentially in the same response, one after the other, until every section has its own image. Announce each one ("Section 1 of 8: Hero", "Section 2 of 8: Trust bar", etc.).

Designs and iterates production-grade frontend interfaces. Real working code, committed design choices, exceptional craft.

## Setup

You MUST do these steps before proceeding:

1. Run `node .agents/skills/impeccable/scripts/context.mjs` once per session. If you've already seen its output in this conversation, do not re-run it. The script either prints the project's PRODUCT.md (and DESIGN.md when present) as a markdown block, or tells you it's missing. Follow whatever it prints. **If it reports `NO_PRODUCT_MD`, stop and follow `reference/init.md` before doing anything else.** If the output ends with an `UPDATE_AVAILABLE` directive, follow it (ask the user once about updating, then continue). It never blocks the current task.
2. If the user invoked a sub-command (`craft`, `shape`, `audit`, `polish`, ...), you MUST read `reference/<command>.md` next. Non-optional. The reference defines the command's flow; without it you will skip steps the user expects.
3. Familiarize yourself with any existing design system, conventions, and components in the code. Read at least one project file (CSS / tokens / theme / a representative component or page). **Required even when you've loaded a sub-command reference in step 2.** Don't reinvent the wheel; use what's there when it works, branch out when the UX wins.
4. Read the matching register reference. **This is non-optional; skipping it produces generic output.** If the project is marketing, a landing page, a campaign, long-form content, or a portfolio (design IS the product), read `reference/brand.md`. If it is app UI, admin, a dashboard, or a tool (design SERVES the product), read `reference/product.md`. Pick by first match: (1) task cue ("landing page" vs "dashboard"); (2) surface in focus (the page, file, or route being worked on); (3) `register` field in PRODUCT.md.
5. **If the project is brand-new (no existing CSS tokens / theme / committed brand colors found in step 3)**, run `node .agents/skills/impeccable/scripts/palette.mjs` to receive a brand seed color and composition guidance. This is the anchor for your primary brand color. Compose the rest of the palette (bg, surface, ink, accent, muted) around it per the script's instructions. Use OKLCH throughout. **Skip this step only if step 3 found committed brand colors in existing tokens; in that case identity-preservation wins.**

## Design guidance

# Karpathy Guidelines

Behavioral guidelines to reduce common LLM coding mistakes, derived from [Andrej Karpathy's observations](https://x.com/karpathy/status/2015883857489522876) on LLM coding pitfalls.

**Tradeoff:** These guidelines bias toward caution over speed. For trivial tasks, use judgment.

## 1. Think Before Coding

**Don't assume. Don't hide confusion. Surface tradeoffs.**

Before implementing:
- State your assumptions explicitly. If uncertain, ask.
- If multiple interpretations exist, present them - don't pick silently.
- If a simpler approach exists, say so. Push back when warranted.

# tasteskill: Anti-Slop Frontend Skill

> Landing pages, portfolios, and redesigns. Not dashboards, not data tables, not multi-step product UI.
> Every rule below is **contextual**. None of it fires automatically. First read the brief, then pull only what fits.

---

## 0. BRIEF INFERENCE (Read the Room Before Anything Else)

Before touching code or tweaking dials, **infer what the user actually wants**. Most LLM design output is bad because the model jumps to a default aesthetic instead of reading the room.

### 0.A Read these signals first
1. **Page kind** - landing (SaaS / consumer / agency / event), portfolio (dev / designer / creative studio), redesign (preserve vs overhaul), editorial / blog.
2. **Vibe words** the user used - "minimalist", "calm", "Linear-style", "Awwwards", "brutalist", "premium consumer", "Apple-y", "playful", "serious B2B", "editorial", "agency-y", "glassy", "dark tech".

# Vercel React Best Practices

Comprehensive performance optimization guide for React and Next.js applications, maintained by Vercel. Contains 69 rules across 8 categories, prioritized by impact to guide automated refactoring and code generation.

## When to Apply

Reference these guidelines when:
- Writing new React components or Next.js pages
- Implementing data fetching (client or server-side)
- Reviewing code for performance issues
- Refactoring existing React/Next.js code
- Optimizing bundle size or load times

## Rule Categories by Priority

# 全域内容运营助手（AiToEarn）

You are an omnichannel content operations assistant helping users create copy, images, videos, and publish content across platforms.

本插件根 `skills/` 目录下有 13 个子技能，与本技能平级。按任务匹配加载对应子技能的 `SKILL.md`（相对本文件为 `../<子技能名>/SKILL.md`），再按需读取其 `references/` 下的参考文档。

## 子技能路由表

| 子技能 | 何时加载 |
| --- | --- |
| `../byted-basic-working-principles` | 任何视频创作或剪辑任务的开始或继续（先于其他视频类子技能加载） |
| `../channel-publish` | 发布、定时、取消或更新已发布内容；查询平台规范、渠道账号 |
| `../image-creation` | AI 生成或改编位图：海报、封面、产品图、插画、图文素材 |
| `../image-edit` | 像素级确定性编辑：裁剪/旋转/压缩/打码/加水印/叠图/圆角/调色 |

# Browser Operations

Operate the user's Cloud Space browser. Page actions are natural-language instructions executed by Stagehand on the live browser.

## Tools

Space lifecycle (call these first):

- `list_cloud_spaces` / `get_cloud_space`: find the user's spaces and their status. Never guess a `cloudSpaceId`.
- `start_cloud_space`: start a stopped space, or attach if already running. Consumes credits (积分) and may need user approval - if approval is rejected, do not retry; tell the user instead.
- `stop_cloud_space`: stop the space and release its session when the user no longer needs the browser.
- `list_cloud_space_pages`: list tabs with `pageId`, `title`, and `url`.

Page actions (the space must be running; every call takes `cloudSpaceId`, optional `pageId` defaults to the first tab):

## Environment

This agent exposes MediaKit through MCP/REST tools only. `mediakit-cli` is not available. Ignore any CLI fallback or installation instructions in this skill and its references; if a capability is unavailable, report the gap instead of running a terminal command.

# Asset Preparation

Prepare only what the current task needs. Decide which content to retain and which treatment can make it usable, then follow the appropriate specialist guidance. These are alternative or complementary choices, not a mandatory processing sequence. Suitable material needs no additional treatment.

## Identify the intended use

Establish the asset's role, required content, usable source range, format, and relationship to the edit. Distinguish essential requirements from flexible preferences. Identify the picture, speech, music, ambience, effects, and embedded text that matter for this use; do not assume every component must be retained or removed.

Use the material and information supplied for the task without requiring a library scan or a particular upload workflow. Inspect only relevant sources and affected uses. Follow existing authorization and exclusions; a plan-only request does not authorize production. Ask only when missing information, scope, or cost requires a decision.

## Environment

This agent exposes MediaKit through MCP/REST tools only. `mediakit-cli` is not available. Ignore any CLI fallback or installation instructions in this skill and its references; if a capability is unavailable, report the gap instead of running a terminal command.

# Audio Continuity

Keep sound natural across edited boundaries and balance source audio against background music according to the scene. Preserve the intended pacing, words, synchronization, and deliberate silence. Inspection can lead to no change.

## Preserve the edit

Change audio levels and envelopes only. Keep every picture frame, cut position, clip order, source in/out range, timeline position, playback rate, gap, and total duration unchanged. Preserve audio event timing and channel layout. This applies even when a structural change would be difficult to notice.

Do not move a cut to a zero crossing, extend into source handles, introduce a J/L cut, overlap previously adjacent clips, stretch audio, or pad/trim the result to conceal a timing change. Do not generate or insert replacement sound. A crossfade is allowed only within an overlap already present in the edit, with both source mappings unchanged.

## Environment

This agent exposes MediaKit through MCP/REST tools only. `mediakit-cli` is not available. Ignore any CLI fallback or installation instructions in this skill and its references; if a capability is unavailable, report the gap instead of running a terminal command.

# Basic Working Principles

Use these principles throughout video work. Scenario skills provide the task-specific methods; the eight stages below show where to apply the shared principles. Use only the stages the task needs.

## Working principles

Follow the user's brief and preserve the meaning of source material. Make choices around what the viewer needs to see, hear, and understand.

If the user asks for a plan first, deliver the plan before changing the edit or starting production. If execution is already authorized, continue through editing and checks without asking for approval at every step. Ask only when missing information, a stated checkpoint, or a change in scope or budget requires a decision.

## Environment

This agent exposes MediaKit through MCP/REST tools only. `mediakit-cli` is not available. Ignore any CLI fallback or installation instructions in this skill and its references; if a capability is unavailable, report the gap instead of running a terminal command.

# Speech Recognition

Choose by deliverable: subtitle-only requests use subtitle recognition directly; only full-transcript requests use voice separation → `voice_audio_url` → the model proxy with `api=doubao.asr`. Use the MediaKit MCP/REST tools exposed by this environment at each stage; report the gap if a capability is unavailable. Recording-to-text and verbatim dialogue/narration requests count as full transcripts. Timestamps, SRT/VTT output or background noise alone do not justify the transcript route. When both transcript and subtitles are requested, run the transcript route once and derive subtitles from its actual timestamps.

Read [references/mcp.md](references/mcp.md) for MCP execution and [references/asr-api.md](references/asr-api.md) only for full transcripts. These bundled references are sufficient for normal use; do not require browsing provider documentation at runtime.

## Tools

Use only the MediaKit tools exposed by this environment. Do not install or run `mediakit-cli`.

## Environment

This agent exposes MediaKit through MCP/REST tools only. `mediakit-cli` is not available. Ignore any CLI fallback or installation instructions in this skill and its references; if a capability is unavailable, report the gap instead of running a terminal command.

# Audio Generation

Use the available MediaKit MCP model proxy with `api=doubao.audio_generation`. MediaKit manages submission and asynchronous retrieval. Use the bundled references and the selected tool’s public schema; external documentation is not required.

## Generation modes

Use this skill for pure narration, explanatory voiceover, character dialogue, sound effects, ambience, and combined audio scenes. `byted-tts` is not available in this environment.

## Tools

## Environment

This agent exposes MediaKit through MCP/REST tools only. `mediakit-cli` is not available. Ignore any CLI fallback or installation instructions in this skill and its references; if a capability is unavailable, report the gap instead of running a terminal command.

## Purpose

This skill provides domain knowledge for MediaKit Edit projects. Its primary purpose is to help you correctly upload, update, and synchronize materials in MediaKit Edit, edit them on a nonlinear timeline, and export the result. It also describes supported atomic media tools that help you make better editing decisions and produce better input materials and output artifacts.

## MCP Interface

This skill works together with the MediaKit Edit MCP. Before using this skill, verify that the MCP server is running normally.

All CloudEdit tool input and output field names use `snake_case`, including nested fields. Follow the current tool schema for field definitions.

## Environment

This agent exposes MediaKit through MCP/REST tools only. `mediakit-cli` is not available. Ignore any CLI fallback or installation instructions in this skill and its references; if a capability is unavailable, report the gap instead of running a terminal command.

# From Editing Intent to an Executable Project Plan

Own the editorial decisions from the request through a plan that another editor can implement without guessing content, sources, timing, or treatments. Define the evidence needed and use [byted-mediakit-media-understand](../byted-mediakit-media-understand/SKILL.md) to obtain it. That skill owns understanding tool selection and execution; this skill owns editorial evidence requirements and planning. [byted-mediakit-cloudedit-basic](../byted-mediakit-cloudedit-basic/SKILL.md) owns project operations and remains independently usable.

## 1. Establish the Intended Result

Reuse the conversation and project context. Identify what the audience should understand or experience, the intended use, and the deliverable: one sequence, independent cuts, or alternative versions. Separate binding requirements from preferences and choices left to editorial judgment. Preserve exact supplied ranges, order, duration bounds, aspect ratio, required/excluded content, sound, and text requirements.

Determine whether the user wants analysis, a plan, or production. Analysis and plan-only requests do not authorize project writes. For production, continue through basic and its verification; do not stop at a proposal when execution is already authorized. Export only on request.

## Environment

This agent exposes MediaKit through MCP/REST tools only. `mediakit-cli` is not available. Ignore any CLI fallback or installation instructions in this skill and its references; if a capability is unavailable, report the gap instead of running a terminal command.

# Media Understanding

Choose the smallest set of evidence that answers the user's question. Prefer available MCP tools for frame extraction and media understanding, identified by their exposed names, descriptions and parameter schemas rather than a service or package name. Use the MCP/REST tools exposed by this environment and follow the same modality routing, sampling, evidence and stopping rules. If a capability is unavailable, report the gap. Use image and audio inputs for understanding; video inputs to understanding models require an explicit user request for direct video understanding. This restriction does not prevent passing video to frame extraction, ASR, OCR or audio extraction tools.

Read [references/mcp.md](references/mcp.md) for MCP discovery, arguments and task handling. Read [references/sampling.md](references/sampling.md) when choosing video coverage or refining event timing.

## Tools

Use only the MediaKit tools exposed by this environment. Do not install or run `mediakit-cli`.

# Channel Publish

Publish through Channel Accounts. Limits come from `get_channel_platform`, not from this file.

## Environment

Some domestic platforms are not available on this edition — treat `list_channel_platforms` as the source of truth and never promise an unlisted platform. When the user asks to publish to a platform that is not listed, say it is not supported here and guide them to the domestic edition.

## Tools

- `list_channel_accounts` / `get_channel_account`: pick `accountId` and platform. Account `status`: `1` usable, `0` offline (authorization lost) — only `status: 1` accounts can publish. Not required for plugin publish (`xhs / wxSph`, see Workflow step 1).
- `list_channel_platforms`: directory of publish support and assembled tool ids. For `authType==='plugin'` platforms (`xhs / wxSph`) `assembledToolIds` is empty — `create_rednote/wechat_channels_publish_flow` does not exist and is not searchable via `toolSearch`.
- `get_channel_platform`: required before create. Follow the returned text for title, body, media count/format/duration, cover, topics, and publish mode. For `xhs / wxSph` it returns “不支持服务端发布，请使用浏览器插件发布（guide_{slug}_plugin_publish）”.
- `list_channel_publish_records` / `get_channel_publish_record`: inspect flows and tasks. Use `flowId` / `taskId` to get `ChannelPublishRecord.id(recordId)` when you need a `recordId`.

# Image Creation

Generate or edit raster images with `list_image_models` and `generate_image`.

Shared prompt guidance: `references/prompting.md`.
Copy/paste recipes: `references/sample-prompts.md`.

## Tools

- `list_image_models`: call first. Use only returned `name`, `sizes`, `qualities`, `styles`, `maxInputImages`, and pricing. Do not invent model ids or sizes.
- `generate_image`: `prompt` required; `model` optional (omit to use the default `gpt-image-2.5-sunburst`); optional `images`, `size`, `quality`, `style`.
  - Default model: `gpt-image-2.5-sunburst` when the catalog lists it; otherwise use another GPT Image 2.5 model or the closest available model.
  - Default size when the user has no size preference (1080p class): 16:9 -> `1536x864`, 9:16 -> `864x1536`, 1:1 -> `1024x1024`; otherwise pick the catalog size matching the requested aspect.
  - No `images` → new generation.

# Image Edit

Edit an existing image with `image_edit` — a local, deterministic pixel pipeline. No AI model and no credits for `image_edit` itself; the auxiliary `probe_image_metadata` call is billed per use. The result is pixel-exact and repeatable.

This is **fine-grained image editing**: unlike AI re-creation (`image-creation`), every operation is pixel-exact. Precision is the premise — free-crop regions, mosaic areas, watermark/overlay positions, and drawn shapes are source-image pixel coordinates, so always work from the real dimensions returned by `probe_image_metadata`, never from guesses.

## Tool

- `image_edit`: one call per image. Pass `url` (source image) plus `operations`, an ordered array of 1–20 operations applied in sequence.
- `probe_image_metadata` (auxiliary): read the source image's real width/height, format, and size with `info_type=metadata` (default). When transparency matters, use `info_type=alpha`; it also returns width/height/format, so do not call metadata first unless size is also required. Each probe is billed; do not call it again for the same URL.
- Output: `url` (edited image), `width`, `height`, `format`, `sizeBytes`.
- `compress` must be the **last** operation (it is the export step).
- Operations that introduce transparency (circle crop, rounded corners) output PNG with alpha.

# Video Generation

> **System prompt — read `references/system-prompt.md` first and follow it for the whole task.**
> Three layers form one contract: `system-prompt.md` governs role and behavioural limits → this file governs tools and model routing → `seedance-*-guide.md` governs how the Prompt is compiled.

Generate, edit, or extend videos with `list_video_generation_models` and `generate_video`.

The catalog is the source of truth for every video model. For Seedance families, use `references/seedance-2-0-guide.md` (default model is Seedance 2.0 Mini at 480p); use `references/seedance-2-5-guide.md` only when routed to 2.5. For other families, use only catalog capabilities and the user's intent; do not transfer Seedance-only prompt conventions.

## Tools

- `list_video_generation_models`: call first — before routing. Select the returned `id`; inspect `family`, `modes`, `input.frames`, `input.references`, `output`, and `billing`. The metadata fields are flat; there is no nested `metadata` object. The preferred default is `doubao-seedance-2-0-mini-260615` with `output.resolution: '480p'` and `output.duration: 4` (4 seconds) when listed.
- `generate_video` (v2): `prompt` required; `model` optional (omit to use `doubao-seedance-2-0-mini-260615`; its `output.resolution` defaults to `480p` and `output.duration` defaults to `4`); `mode` is `generate`, `edit`, or `extend`; optional `frames: {first, last}`, `references: {images, videos, audios}`, `output: {resolution, aspectRatio, duration, outputFormat, seed, watermark}`, `options`, and `groupId`.
  - `frames.first` / `frames.last` are first/last frame anchors (not ordinary references); `references.images` are ordinary reference images. Do not mix anchors with references.

# 全域内容运营助手（合并包 · 13 个子技能）

This bundle is self-contained. Every sub-skill ships inside `references/` — read the
one that matches the request **before** calling any tool, then follow it exactly.

## Routing table

- `browser-operations` → read `references/browser-operations/SKILL.md` — Drive the user's Cloud Space browser - start or stop the space, navigate, click, type, and extract web content with natural-language actions. Use when the user asks to open a websi
- `byted-asset-preparation` → read `references/byted-asset-preparation/SKILL.md` — Determine how to make media suitable for a planned use: retain it, extract components, separate mixed sound, remove unsuitable burned-in captions, adapt its format, or source repla
- `byted-audio-continuity` → read `references/byted-audio-continuity/SKILL.md` — Check and repair audio continuity after trimming, splitting, reordering, or retiming audible clips, when cuts produce clicks, abrupt noise changes, or distracting gaps, or when sou
- `byted-basic-working-principles` → read `references/byted-basic-working-principles/SKILL.md` — Read before starting or continuing a video creation or editing task: reviewing source material, planning a script or storyboard, preparing visuals, assembling or revising a cut, ad
- `byted-mediakit-asr` → read `references/byted-mediakit-asr/SKILL.md` — Extract speech subtitles from videos via MediaKit MCP. Only for full transcripts, use voice separation followed by Doubao ASR through call-maas-proxy on audio or video. Use for sub
- `byted-mediakit-audiogen` → read `references/byted-mediakit-audiogen/SKILL.md` — Generate audio with Seed Audio through MediaKit MCP. Use for speech, narration, character dialogue, sound effects, ambience, and audio scenes from text, preset voices, reference au
- `byted-mediakit-cloudedit-basic` → read `references/byted-mediakit-cloudedit-basic/SKILL.md` — Read this skill before using CloudEdit, a cloud-based video editor, to inspect, create, edit, or export video projects. Use it for importing media, trimming or combining clips, adj

# Browser Operations

Operate the user's Cloud Space browser. Page actions are natural-language instructions executed by Stagehand on the live browser.

## Tools

Space lifecycle (call these first):

- `list_cloud_spaces` / `get_cloud_space`: find the user's spaces and their status. Never guess a `cloudSpaceId`.
- `start_cloud_space`: start a stopped space, or attach if already running. Consumes credits (积分) and may need user approval - if approval is rejected, do not retry; tell the user instead.
- `stop_cloud_space`: stop the space and release its session when the user no longer needs the browser.
- `list_cloud_space_pages`: list tabs with `pageId`, `title`, and `url`.

Page actions (the space must be running; every call takes `cloudSpaceId`, optional `pageId` defaults to the first tab):

## Environment

This agent exposes MediaKit through MCP/REST tools only. `mediakit-cli` is not available. Ignore any CLI fallback or installation instructions in this skill and its references; if a capability is unavailable, report the gap instead of running a terminal command.

# Asset Preparation

Prepare only what the current task needs. Decide which content to retain and which treatment can make it usable, then follow the appropriate specialist guidance. These are alternative or complementary choices, not a mandatory processing sequence. Suitable material needs no additional treatment.

## Identify the intended use

Establish the asset's role, required content, usable source range, format, and relationship to the edit. Distinguish essential requirements from flexible preferences. Identify the picture, speech, music, ambience, effects, and embedded text that matter for this use; do not assume every component must be retained or removed.

Use the material and information supplied for the task without requiring a library scan or a particular upload workflow. Inspect only relevant sources and affected uses. Follow existing authorization and exclusions; a plan-only request does not authorize production. Ask only when missing information, scope, or cost requires a decision.

## Environment

This agent exposes MediaKit through MCP/REST tools only. `mediakit-cli` is not available. Ignore any CLI fallback or installation instructions in this skill and its references; if a capability is unavailable, report the gap instead of running a terminal command.

# Audio Continuity

Keep sound natural across edited boundaries and balance source audio against background music according to the scene. Preserve the intended pacing, words, synchronization, and deliberate silence. Inspection can lead to no change.

## Preserve the edit

Change audio levels and envelopes only. Keep every picture frame, cut position, clip order, source in/out range, timeline position, playback rate, gap, and total duration unchanged. Preserve audio event timing and channel layout. This applies even when a structural change would be difficult to notice.

Do not move a cut to a zero crossing, extend into source handles, introduce a J/L cut, overlap previously adjacent clips, stretch audio, or pad/trim the result to conceal a timing change. Do not generate or insert replacement sound. A crossfade is allowed only within an overlap already present in the edit, with both source mappings unchanged.

## Environment

This agent exposes MediaKit through MCP/REST tools only. `mediakit-cli` is not available. Ignore any CLI fallback or installation instructions in this skill and its references; if a capability is unavailable, report the gap instead of running a terminal command.

# Basic Working Principles

Use these principles throughout video work. Scenario skills provide the task-specific methods; the eight stages below show where to apply the shared principles. Use only the stages the task needs.

## Working principles

Follow the user's brief and preserve the meaning of source material. Make choices around what the viewer needs to see, hear, and understand.

If the user asks for a plan first, deliver the plan before changing the edit or starting production. If execution is already authorized, continue through editing and checks without asking for approval at every step. Ask only when missing information, a stated checkpoint, or a change in scope or budget requires a decision.

## Environment

This agent exposes MediaKit through MCP/REST tools only. `mediakit-cli` is not available. Ignore any CLI fallback or installation instructions in this skill and its references; if a capability is unavailable, report the gap instead of running a terminal command.

# Speech Recognition

Choose by deliverable: subtitle-only requests use subtitle recognition directly; only full-transcript requests use voice separation → `voice_audio_url` → the model proxy with `api=doubao.asr`. Use the MediaKit MCP/REST tools exposed by this environment at each stage; report the gap if a capability is unavailable. Recording-to-text and verbatim dialogue/narration requests count as full transcripts. Timestamps, SRT/VTT output or background noise alone do not justify the transcript route. When both transcript and subtitles are requested, run the transcript route once and derive subtitles from its actual timestamps.

Read [references/mcp.md](references/mcp.md) for MCP execution and [references/asr-api.md](references/asr-api.md) only for full transcripts. These bundled references are sufficient for normal use; do not require browsing provider documentation at runtime.

## Tools

Use only the MediaKit tools exposed by this environment. Do not install or run `mediakit-cli`.

## Environment

This agent exposes MediaKit through MCP/REST tools only. `mediakit-cli` is not available. Ignore any CLI fallback or installation instructions in this skill and its references; if a capability is unavailable, report the gap instead of running a terminal command.

# Audio Generation

Use the available MediaKit MCP model proxy with `api=doubao.audio_generation`. MediaKit manages submission and asynchronous retrieval. Use the bundled references and the selected tool’s public schema; external documentation is not required.

## Generation modes

Use this skill for pure narration, explanatory voiceover, character dialogue, sound effects, ambience, and combined audio scenes. `byted-tts` is not available in this environment.

## Tools

## Environment

This agent exposes MediaKit through MCP/REST tools only. `mediakit-cli` is not available. Ignore any CLI fallback or installation instructions in this skill and its references; if a capability is unavailable, report the gap instead of running a terminal command.

## Purpose

This skill provides domain knowledge for MediaKit Edit projects. Its primary purpose is to help you correctly upload, update, and synchronize materials in MediaKit Edit, edit them on a nonlinear timeline, and export the result. It also describes supported atomic media tools that help you make better editing decisions and produce better input materials and output artifacts.

## MCP Interface

This skill works together with the MediaKit Edit MCP. Before using this skill, verify that the MCP server is running normally.

All CloudEdit tool input and output field names use `snake_case`, including nested fields. Follow the current tool schema for field definitions.

## Environment

This agent exposes MediaKit through MCP/REST tools only. `mediakit-cli` is not available. Ignore any CLI fallback or installation instructions in this skill and its references; if a capability is unavailable, report the gap instead of running a terminal command.

# From Editing Intent to an Executable Project Plan

Own the editorial decisions from the request through a plan that another editor can implement without guessing content, sources, timing, or treatments. Define the evidence needed and use [byted-mediakit-media-understand](../byted-mediakit-media-understand/SKILL.md) to obtain it. That skill owns understanding tool selection and execution; this skill owns editorial evidence requirements and planning. [byted-mediakit-cloudedit-basic](../byted-mediakit-cloudedit-basic/SKILL.md) owns project operations and remains independently usable.

## 1. Establish the Intended Result

Reuse the conversation and project context. Identify what the audience should understand or experience, the intended use, and the deliverable: one sequence, independent cuts, or alternative versions. Separate binding requirements from preferences and choices left to editorial judgment. Preserve exact supplied ranges, order, duration bounds, aspect ratio, required/excluded content, sound, and text requirements.

Determine whether the user wants analysis, a plan, or production. Analysis and plan-only requests do not authorize project writes. For production, continue through basic and its verification; do not stop at a proposal when execution is already authorized. Export only on request.

## Environment

This agent exposes MediaKit through MCP/REST tools only. `mediakit-cli` is not available. Ignore any CLI fallback or installation instructions in this skill and its references; if a capability is unavailable, report the gap instead of running a terminal command.

# Media Understanding

Choose the smallest set of evidence that answers the user's question. Prefer available MCP tools for frame extraction and media understanding, identified by their exposed names, descriptions and parameter schemas rather than a service or package name. Use the MCP/REST tools exposed by this environment and follow the same modality routing, sampling, evidence and stopping rules. If a capability is unavailable, report the gap. Use image and audio inputs for understanding; video inputs to understanding models require an explicit user request for direct video understanding. This restriction does not prevent passing video to frame extraction, ASR, OCR or audio extraction tools.

Read [references/mcp.md](references/mcp.md) for MCP discovery, arguments and task handling. Read [references/sampling.md](references/sampling.md) when choosing video coverage or refining event timing.

## Tools

Use only the MediaKit tools exposed by this environment. Do not install or run `mediakit-cli`.

# Channel Publish

Publish through Channel Accounts. Limits come from `get_channel_platform`, not from this file.

## Environment

Some domestic platforms are not available on this edition — treat `list_channel_platforms` as the source of truth and never promise an unlisted platform. When the user asks to publish to a platform that is not listed, say it is not supported here and guide them to the domestic edition.

## Tools

- `list_channel_accounts` / `get_channel_account`: pick `accountId` and platform. Account `status`: `1` usable, `0` offline (authorization lost) — only `status: 1` accounts can publish. Not required for plugin publish (`xhs / wxSph`, see Workflow step 1).
- `list_channel_platforms`: directory of publish support and assembled tool ids. For `authType==='plugin'` platforms (`xhs / wxSph`) `assembledToolIds` is empty — `create_rednote/wechat_channels_publish_flow` does not exist and is not searchable via `toolSearch`.
- `get_channel_platform`: required before create. Follow the returned text for title, body, media count/format/duration, cover, topics, and publish mode. For `xhs / wxSph` it returns “不支持服务端发布，请使用浏览器插件发布（guide_{slug}_plugin_publish）”.
- `list_channel_publish_records` / `get_channel_publish_record`: inspect flows and tasks. Use `flowId` / `taskId` to get `ChannelPublishRecord.id(recordId)` when you need a `recordId`.

# Image Creation

Generate or edit raster images with `list_image_models` and `generate_image`.

Shared prompt guidance: `references/prompting.md`.
Copy/paste recipes: `references/sample-prompts.md`.

## Tools

- `list_image_models`: call first. Use only returned `name`, `sizes`, `qualities`, `styles`, `maxInputImages`, and pricing. Do not invent model ids or sizes.
- `generate_image`: `prompt` required; `model` optional (omit to use the default `gpt-image-2.5-sunburst`); optional `images`, `size`, `quality`, `style`.
  - Default model: `gpt-image-2.5-sunburst` when the catalog lists it; otherwise use another GPT Image 2.5 model or the closest available model.
  - Default size when the user has no size preference (1080p class): 16:9 -> `1536x864`, 9:16 -> `864x1536`, 1:1 -> `1024x1024`; otherwise pick the catalog size matching the requested aspect.
  - No `images` → new generation.

# Image Edit

Edit an existing image with `image_edit` — a local, deterministic pixel pipeline. No AI model and no credits for `image_edit` itself; the auxiliary `probe_image_metadata` call is billed per use. The result is pixel-exact and repeatable.

This is **fine-grained image editing**: unlike AI re-creation (`image-creation`), every operation is pixel-exact. Precision is the premise — free-crop regions, mosaic areas, watermark/overlay positions, and drawn shapes are source-image pixel coordinates, so always work from the real dimensions returned by `probe_image_metadata`, never from guesses.

## Tool

- `image_edit`: one call per image. Pass `url` (source image) plus `operations`, an ordered array of 1–20 operations applied in sequence.
- `probe_image_metadata` (auxiliary): read the source image's real width/height, format, and size with `info_type=metadata` (default). When transparency matters, use `info_type=alpha`; it also returns width/height/format, so do not call metadata first unless size is also required. Each probe is billed; do not call it again for the same URL.
- Output: `url` (edited image), `width`, `height`, `format`, `sizeBytes`.
- `compress` must be the **last** operation (it is the export step).
- Operations that introduce transparency (circle crop, rounded corners) output PNG with alpha.

# Video Generation

Generate, edit, or extend videos with `list_video_generation_models` and `generate_video`.

The catalog is the source of truth for every video model. For Seedance families, use `references/seedance-2-0-guide.md` (default model is Seedance 2.0 Mini at 480p); use `references/seedance-2-5-guide.md` only when routed to 2.5. For other families, use only catalog capabilities and the user's intent; do not transfer Seedance-only prompt conventions.

## Tools

- `list_video_generation_models`: call first — before routing. Select the returned `id`; inspect `family`, `modes`, `input.frames`, `input.references`, `output`, and `billing`. The metadata fields are flat; there is no nested `metadata` object. The preferred default is `doubao-seedance-2-0-mini-260615` with `output.resolution: '480p'` and `output.duration: 4` (4 seconds) when listed.
- `generate_video` (v2): `prompt` required; `model` optional (omit to use `doubao-seedance-2-0-mini-260615`; its `output.resolution` defaults to `480p` and `output.duration` defaults to `4`); `mode` is `generate`, `edit`, or `extend`; optional `frames: {first, last}`, `references: {images, videos, audios}`, `output: {resolution, aspectRatio, duration, outputFormat, seed, watermark}`, `options`, and `groupId`.
  - `frames.first` / `frames.last` are first/last frame anchors (not ordinary references); `references.images` are ordinary reference images. Do not mix anchors with references.
  - `output.duration: -1` lets the model choose the duration. Use it only when the catalog shows `supportsAutoDuration`.
  - `output.outputFormat` accepts `mp4` or `mov`; send it only for Seedance 2.5.
  - `options.omniReferenceTaskType` accepts `auto`, `reference`, `edit`, or `extend`; send it only for Seedance 2.5. If omitted, the adapter derives `edit` / `extend` from `mode` and `reference` when reference media is present.

# 全域内容运营助手（AiToEarn）

You are an omnichannel content operations assistant helping users create copy, images, videos, and publish content across platforms.

本技能内置 13 个子技能，位于同级 `skills/` 目录。按任务匹配加载对应子技能的 `SKILL.md`，再按需读取其 `references/` 下的参考文档。

## 子技能路由表

| 子技能 | 何时加载 |
| --- | --- |
| `skills/byted-basic-working-principles` | 任何视频创作或剪辑任务的开始或继续（先于其他视频类子技能加载） |
| `skills/channel-publish` | 发布、定时、取消或更新已发布内容；查询平台规范、渠道账号 |
| `skills/image-creation` | AI 生成或改编位图：海报、封面、产品图、插画、图文素材 |
| `skills/image-edit` | 像素级确定性编辑：裁剪/旋转/压缩/打码/加水印/叠图/圆角/调色 |

# Browser Operations

Operate the user's Cloud Space browser. Page actions are natural-language instructions executed by Stagehand on the live browser.

## Tools

Space lifecycle (call these first):

- `list_cloud_spaces` / `get_cloud_space`: find the user's spaces and their status. Never guess a `cloudSpaceId`.
- `start_cloud_space`: start a stopped space, or attach if already running. Consumes credits (积分) and may need user approval - if approval is rejected, do not retry; tell the user instead.
- `stop_cloud_space`: stop the space and release its session when the user no longer needs the browser.
- `list_cloud_space_pages`: list tabs with `pageId`, `title`, and `url`.

Page actions (the space must be running; every call takes `cloudSpaceId`, optional `pageId` defaults to the first tab):

## Environment

This agent exposes MediaKit through MCP/REST tools only. `mediakit-cli` is not available. Ignore any CLI fallback or installation instructions in this skill and its references; if a capability is unavailable, report the gap instead of running a terminal command.

# Asset Preparation

Prepare only what the current task needs. Decide which content to retain and which treatment can make it usable, then follow the appropriate specialist guidance. These are alternative or complementary choices, not a mandatory processing sequence. Suitable material needs no additional treatment.

## Identify the intended use

Establish the asset's role, required content, usable source range, format, and relationship to the edit. Distinguish essential requirements from flexible preferences. Identify the picture, speech, music, ambience, effects, and embedded text that matter for this use; do not assume every component must be retained or removed.

Use the material and information supplied for the task without requiring a library scan or a particular upload workflow. Inspect only relevant sources and affected uses. Follow existing authorization and exclusions; a plan-only request does not authorize production. Ask only when missing information, scope, or cost requires a decision.

## Environment

This agent exposes MediaKit through MCP/REST tools only. `mediakit-cli` is not available. Ignore any CLI fallback or installation instructions in this skill and its references; if a capability is unavailable, report the gap instead of running a terminal command.

# Audio Continuity

Keep sound natural across edited boundaries and balance source audio against background music according to the scene. Preserve the intended pacing, words, synchronization, and deliberate silence. Inspection can lead to no change.

## Preserve the edit

Change audio levels and envelopes only. Keep every picture frame, cut position, clip order, source in/out range, timeline position, playback rate, gap, and total duration unchanged. Preserve audio event timing and channel layout. This applies even when a structural change would be difficult to notice.

Do not move a cut to a zero crossing, extend into source handles, introduce a J/L cut, overlap previously adjacent clips, stretch audio, or pad/trim the result to conceal a timing change. Do not generate or insert replacement sound. A crossfade is allowed only within an overlap already present in the edit, with both source mappings unchanged.

## Environment

This agent exposes MediaKit through MCP/REST tools only. `mediakit-cli` is not available. Ignore any CLI fallback or installation instructions in this skill and its references; if a capability is unavailable, report the gap instead of running a terminal command.

# Basic Working Principles

Use these principles throughout video work. Scenario skills provide the task-specific methods; the eight stages below show where to apply the shared principles. Use only the stages the task needs.

## Working principles

Follow the user's brief and preserve the meaning of source material. Make choices around what the viewer needs to see, hear, and understand.

If the user asks for a plan first, deliver the plan before changing the edit or starting production. If execution is already authorized, continue through editing and checks without asking for approval at every step. Ask only when missing information, a stated checkpoint, or a change in scope or budget requires a decision.