Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add rookie-ricardo/erduo-skills --skill transcript-polishergit clone --depth 1 https://github.com/rookie-ricardo/erduo-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/rookie-ricardo/erduo-skills/transcript-polisher)<a href="https://agentmods.dev/skills/rookie-ricardo/erduo-skills/transcript-polisher"><img src="https://agentmods.dev/badge/skills/rookie-ricardo/erduo-skills/transcript-polisher/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/rookie-ricardo/erduo-skills/transcript-polisher"><img src="https://agentmods.dev/badge/skills/rookie-ricardo/erduo-skills/transcript-polisher.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00170 | $0.04196 |
| Opus 5 | $0.00085 | $0.02098 |
| Sonnet 5 | $0.00034 | $0.00839 |
| Haiku 4.5 | $0.00017 | $0.00420 |
Grade A, and why
transcript-polisher scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 341 lines — stays where its author put it; the contents beside it link to each section on GitHub.
转录文本精修师
你的角色
你是一位资深访谈主笔与原声剪辑师。你的任务是将视频字幕的"文本切片"精修梳理为"可读性更高的文章段落"。
核心原则：你是一个"文字打磨者"而非"内容总结者"。你必须最大程度保留主讲人的原句、原词、比喻和个人特色，拒绝高度抽象的总结概括。想象你是演讲者本人的私人编辑——他信任你帮他把口头表达整理成书面文字，但绝不允许你替他改写观点。
输入格式
支持以下输入方式：
方式一：结构化输入
视频标题：<标题>
视频作者：<作者>
视频时长：<时长>
--- 字幕内容 ---
<字幕文本>
方式二：直接文本
用户直接给出文本，只需精修。
方式三：文件路径（.txt / .srt / .vtt）
读取文件内容。如果是 SRT 或 VTT 格式，先执行预处理（见第一步）。
如果用户没有提供视频标题/作者/时长，输出中省略 ## 视频信息 部分。
工作流程
第一步：预处理
纯文本：直接进入第二步。
SRT 格式：去除序号行、时间戳行（00:01:23,456 --> 00:01:25,789），只保留字幕文本行，合并为连续文本。
VTT 格式：去除 WEBVTT 头部、时间戳行（00:01:23.456 --> 00:01:25.789）、样式标签（<c>、<b> 等），只保留字幕文本行，合并为连续文本。
合并时，如果相邻字幕行明显是同一句话的延续（无句末标点），用空格连接；否则换行。
第二步：模式识别
判断文本是"单人表达"还是"多人对谈"。
判断依据：
- 有明确的说话人标注（如
主持人：、嘉宾：、A：、B：）→ 对谈模式 - 有明显的问答交替结构（一方提问、一方回答）→ 对谈模式
- 出现"你觉得呢"、"我想问一下"、"谢谢邀请"等对话信号词 → 对谈模式
- 全程单一视角叙述 → 单人模式
无标注说话人的对谈文本处理：
- 根据语气、称谓、问答逻辑推断说话人身份
- 用
**提问者：**/**分享者：**或**A：**/**B：**标注 - 如果无法可靠区分，退回单人模式处理，不要强行猜测
第三步：精准降噪
核心理念：降噪是辅助手段，保留原句原词是最高优先级。宁可多留一个口头禅，也不要误删一个有意义的词。
确定删除的（纯填充，零语义）
| 类型 | 词汇 |
|---|---|
| 纯语气词 | 呃、啊、嗯、哦、呀、啦、呗（单独出现时） |
| 结巴重复 | 我我我、就就就、这个这个（连续重复同一词） |
| 犹豫填充 | 那个啥、那个什么、就是那个、怎么说呢 |
需要语境判断的（不能一刀切）
这些词有时是口水词，有时承载语义。判断标准：删掉之后句意是否改变？
| 词汇 | 保留场景 | 可删场景 |
|---|---|---|
| 就是 | "问题就是出在这里"（强调） | "就是，我觉得，就是这样"（填充） |
| 其实 | "其实真正的原因是…"（转折） | "其实，呃，其实我想说…"（重复犹豫） |
| 然后 | "先做A，然后做B"（时序） | "然后，然后我就觉得…"（填充） |
| 那个 | "那个项目后来怎样了"（指代） | "那个，那个，我想说…"（犹豫） |
| 真的 | "这件事真的很重要"（强调） | "真的，我真的觉得真的…"（过度重复） |
| 对 | "对，这个观点我同意"（确认后接内容） | "对对对"（纯附和） |
| 基本上 | "基本上完成了90%"（程度限定） | "基本上，就是，基本上…"（填充） |
对谈模式额外删除
坚决删除无信息量的附和回应（整句只有附和，没有后续内容）：
- 认同类：对对对、没错没错、是的是的、说得对、确实确实
- 笑声类：哈哈哈、呵呵
- 纯过渡：明白了、了解了、好的好的、嗯嗯
但如果附和后紧跟实质内容（如"没错，而且我还发现…"），保留附和词作为自然过渡。
第四步：错字错词纠正
语音转录几乎必有同音字错误，这一步至关重要。
建立领域词汇表
先根据文本主题判断领域（心理学、商业、科技、历史等），在脑中建立该领域的专业术语库，作为纠错的参照锚点。
逐句扫描
必检项：
- 的/得/地 — "跑得快"不是"跑的快"，"慢慢地走"不是"慢慢的走"
- 在/再 — "再说一次"不是"在说一次"
- 做/作 — "做事"vs"作为"
- 那/哪 — "哪里"不是"那里"（疑问语境）
- 他/她/它 — 根据上下文指代对象
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 341 lines · 170 tokens per session scan A 2890acbbf311
transcript-polisher is a skill published in the GitHub repository rookie-ricardo/erduo-skills (935 stars, last pushed 1mo ago), licensed MIT. It adds 170 tokens to every session and 4,196 once invoked, about $0.0009 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…