personal writing voice(chinese-writing for a try)

    1

    A Chinese-language writing system that transfers stylistic mechanisms from high-quality sources while protecting your voice.

    Free

    1 useSecurity scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    personal writing voice(chinese-writing for a try)

    Example session with this skill installed

    使用发布模式 (publish) 为我写一篇关于“独立开发者如何保持专注”的初稿。作者身份:cc,风格:wechat-reflection。我已经准备好了材料,请先理解我的核心观点并检查材料风险。

    • Read your context and instructions
    • Compiled the personal writing voice

    我已经理解了您的核心创作意图,并完成了初步的材料审核与风格配置。

    任务预检报告 (Task Preflight)

    检查项状态详细说明
    核心立场已确认强调“专注不是排除干扰,而是建立与目标的深度链接”。
    材料清单良好包含3条个人案例;缺少1处关于“专注时长”的具体量化数据(已标记为不可捏造)。
    模式与风格publish激活 wechat-reflection 风格,隔离其他技术文档样式。
    外部来源待筛选将检索 2016 年前的中文原创深度报道,重点提取其叙事节奏。
    可转移特征规划中预计提取 3-4 个语法机制,包括长短句交替与特定转折逻辑。

    决策请求:
    关于“开发者在深夜工作的专注效率”这一段,您是否拥有具体的实验数据,还是仅作为感性描述?如果缺乏数据,我将在初稿中保持留白或使用模糊处理,以符合您的事实边界。

    Next steps

    1. 确认上述核心立场是否准确。
    2. 决定是否需要我执行 python scripts/pwv.py prepare-task 来锁定当前材料。
    3. 授权我开始搜索符合 zh_original 验证的中文参考资源。

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Writing tools often force a generic "AI voice" on your content, stripping away your personal authority and style. This skill functions as a Chinese influence-research and writing-transfer system designed to preserve your specific judgment while analyzing high-quality Chinese sources for linguistic mechanisms. It treats external articles only as evidence for expression, never as a source for your opinions or facts.

    What it does

    • Voice preservation protects the author's real judgment and material from being overridden by external influence.
    • Source screening verifies Chinese-original articles to ensure style models are not built from SEO farms or AI-generated clutter.
    • Mechanism transfer extracts specific lexical choices and rhythms from reference texts without copying signature wording or metaphors.
    • Material ledgering strictly separates what you know from what the AI is forbidden to invent.
    • Multi-layer gates runs every draft through material, source, style, and close-imitation checks before finalizing.

    How it works

    1. Prepare-Task establishes the author's core judgment, style selection, and material boundaries using the prepare-task command.
    2. Influence Research searches and screens Chinese-original sources to build evidence-backed transfer plans.
    3. Drafting creates a "thought draft" focused on truth and logic before applying any language transfer mechanisms.
    4. Review executes check-draft to run editorial, reader, and privacy gates on the stable candidate.

    Frameworks & tools

    This skill utilizes a Python-based CLI (scripts/pwv.py) for file-backed workflow management and evidence recording. It relies on Markdown-based reference libraries for style storage, influence analysis, and memory gates.

    Why this beats prompting it yourself

    Basic prompting usually results in the AI mimicking a source too closely or hallucinating facts to fit a style. This skill enforces a "Fixed Priority" hierarchy that puts your judgment above all else, using a four-layer fingerprinting system to ensure style transfer is evidence-based and safe from plagiarism.

    Use cases

    • Drafting long-form WeChat public account articles with a specific professional rhythm.
    • Preparing formal Chinese speeches that require high-intensity stylistic restraint.
    • De-AIing existing drafts to restore a natural, irregular human cadence.
    • Managing a private long-term memory of your unique writing rules and feedback.

    Known limitations

    The system stops if publish mode sources cannot be verified as original Chinese content. It requires a local CLI setup for full evidence recording and gate enforcement.

    Requires Python for CLI gate enforcement. Stop-blocks if sources are suspected AI or SEO-farmed. Browsing is mandatory for publish-mode source gates.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    1 use

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 29 days ago

    • Free with an Agensi account

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Environment Variables
    Listed29 days ago

    What's inside

    Frequently Asked Questions

    Unlike standard prompts that often lead to plagiarism or 'AI-sounding' Chinese, this skill uses a four-layer gating system to isolate a source's rhythm and lexical patterns from your actual facts and opinions.

    The skill includes a Python-based CLI (`pwv.py`) and a Markdown-based reference library system to manage your style memory and evidence recording locally.

    This system is specifically optimized for original Chinese-language content and will automatically halt if it detects sources from SEO farms or low-quality AI-generated translations.

    To ensure data privacy and rigorous evidence recording, this skill requires a local CLI setup for file-backed workflow management rather than a simple web-based chat interface.

    By using the `prepare-task` and `check-draft` commands, the system enforces a 'Fixed Priority' hierarchy that treats your specific judgment as the immutable foundation of every piece.

    New in Writing & Content