Transcript Cleanup — AI skill for Claude Code
{{include:common}} 你是會議逐字稿清稿員.
How to install Transcript Cleanup
Installs to ~/.claude/skills/yazelin-mori-canvas-transcript-cleanup/SKILL.md
mkdir -p ~/.claude/skills/yazelin-mori-canvas-transcript-cleanup && curl -fsSL https://raw.githubusercontent.com/yazelin/mori-canvas/HEAD/prompts/transcript-cleanup.md -o ~/.claude/skills/yazelin-mori-canvas-transcript-cleanup/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Transcript Cleanup does
{{include:common}}
你是會議逐字稿清稿員。輸入是語音辨識(STT)的原始輸出,常有錯字、斷錯句、口語贅字。把它清成乾淨的會議記錄文字,規則:
- 修錯字:同音字、相近詞(例:「預月」→「預約」)。
- 重新斷句:依語意補標點(逗號、句號、問號),把黏在一起的句子切開、把被切碎的句子接回。
- 刪贅字冗詞:嗯、呃、欸、那個、就是說、然後然後、對對對、這樣子、的部分、的動作…等口語填充詞;重複的詞只留一次。
- 不增不減語意:不改寫、不縮寫、不擴寫、不加入原文沒有的內容;專有名詞、人名、數字、金額原樣保留。
- 整段都是無意義的語助詞或雜音時輸出空字串。
只輸出清好的繁體中文純文字,不要解釋、不要前言、不要 markdown 圍欄。
Alternatives in AI
- Humanizer-zh: 去除 AI 写作痕迹 — 你是一位文字编辑,专门识别和去除 AI 生成文本的痕迹,使文字听起来更自然、更有人味 5k ★
- Paper Humanize — Light AI-ism cleanup on a paragraph or section 31 ★
- Council Cleanup — Clean up the LLM Council working directory and temporary files created in .council/ 20 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Whisperflow Local
A 100% local WhisperFlow clone for Windows — Ctrl+Win to dictate, faster-whisper (CUDA) + a local Ollama clean
Garbage Man
Provider-neutral email review and future-mail cleanup for AI agents
Agents In A Chest.Nvim
Neovim plugin to run LLM coding agents in isolated Git work trees with auto‑commit, tabbed sessions, and safe
System Cleanup
Agent Skill: safely clean Windows with any AI coding agent (Trae / Claude Code / Cursor / CodeBuddy / ZCode) -
C Disk Cleanup
一个 Claude Code 技能,用 PowerShell 并行扫描 C 盘 11 个关键区域,AI 为每个目录/文件生成中文描述和风险定级(安全 / 需确认 / 危险),输出单文件只读 HTML 报告。
Instruction Health Skills
Agent skills to keep AI instruction files (AGENTS.md, CLAUDE.md, MEMORY.md) lean. Guardian runs before edits t
Related Agents
Context Refinement
Updates task context manifest with discoveries from current work session. Reads transcript to understand what
Session Context
Loads a single FullStory session's event transcript into an isolated context window and answers a specific tas
Skill Applied Judge
Use this agent at session-end Phase 3.6.6 (#645 L3) to judge — from the session transcript tail — whether each