course notes

    1

    Converts PPT and PDF courseware into structured Markdown notes following the original teaching sequence.

    Free

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    course notes

    Example session with this skill installed

    Convert the attached PDF for Chapter 4: Neural Networks into Chinese Markdown notes. Keep all original diagrams and math formulas. Save it to chapter_04_notes.md.

    • Read your context and instructions
    • Compiled the course notes

    I have processed the courseware for Chapter 4: Neural Networks. The notes follow the original teaching sequence, including all definitions for backpropagation and activation functions. I have preserved the original figures as image references and formatted all derivations using LaTeX.

    Chapter 4: Neural Networks (课程笔记)

    4.1 神经元模型 (Neuron Model)

    神经网络最基本的成分是神经元模型。在 M-P 神经元模型中,神经元接收到来自 $n$ 个其他神经元传递过来的输入信号。

    组成部分说明
    输入信号$x_i$
    权重$w_i$
    阈值$\theta$

    激活函数通常使用 Sigmoid 函数:
    $$f(x) = \frac{1}{1 + e^{-x}}$$

    4.2 多层网络与反向传播 (Backpropagation)

    • 误差逆传播算法 (BP): 是目前最成功的神经网络学习算法。
    • 目标: 最小化训练集上的累积误差 $E = \frac{1}{m} \sum_{k=1}^m E_k$。


    图 4.1: 标准前馈神经网络结构

    算法步骤:

    1. 将输入层送入网络,计算各层输出。
    2. 计算输出层的误差。
    3. 将误差逆向传播至隐含层。
    4. 根据误差调节连接权和阈值。

    Next steps

    • 运行 python3 scripts/check_notes.py chapter_04_notes.md 验证图片路径。
    • 如果需要针对梯度消失问题增加额外解释,请告诉我。
    • 确认是否需要将此章节合并到主课程文件中。

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

    About this skill

    Courseware management often leaves students with a fragmented mess of slides, PDFs, and disorganized screenshots. This skill transforms raw PPT, PPTX, and PDF materials into structured, readable Markdown notes that mirror your instructor's original teaching sequence.

    What it does

    • Structure Extraction parses courseware to identify titles, hierarchy, and knowledge points without losing the original logic.
    • Visual Preservation keeps original figures, tables, and complex formulas instead of replacing them with generic text.
    • Content Consolidation merges repetitive slide content while retaining critical definitions, theorems, and derivations.
    • Selective Formatting defaults to Typora-compatible Markdown with support for LaTeX formulas and code fences.
    • Incremental Updates allows you to supplement existing note files or update specific chapters without overwriting your personal annotations.

    How it works

    1. Inventory Materials by providing the PDF or PPTX files and specifying the chapter scope or target output file.
    2. Configure Depth by defining how examples should be treated and whether to include administrative or exam-related notices.
    3. Generate Notes as the agent parses text and visual elements to build a hierarchical Markdown document.
    4. Verify Coverage via internal checks that compare the output against source page ranges to ensure no knowledge points were skipped.

    Frameworks & tools

    This skill works with Markdown, LaTeX, and Python 3 for link verification. It is optimized for the Typora editor but compatible with any Markdown-based environment like Obsidian or VS Code.

    Why this beats prompting it yourself

    Generic prompts often skip "minor" details or hallucinate simplified explanations that miss the technical nuances required for university-level coursework. This skill enforces strict sequence retention and visual object preservation, ensuring your notes actually reflect the source material rather than a summarized approximation.

    Use cases

    • Lecture Consolidation to turn a 60-slide PPT deck into a concise, readable study guide.
    • Note Supplementation for adding a missing chapter to your existing semester notes while keeping your styles consistent.
    • Legacy Conversion to move old PPT courseware into a modern Markdown-based personal knowledge management system.
    • Formula Digitization to extract complex mathematical derivations into properly formatted LaTeX blocks.

    Known limitations

    Does not support automatic textbook restructuring or exam prediction by default. Visual inspection is required for scans; text extraction for navigation may fail on low-quality image-only PDFs.

    Requires manual visual check for low-quality scans. Does not proactively generate exam predictions or textbook-style restructuring. Best suited for Typora-compatible Markdown environments.

    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

    Fresh listing

    Recently published to Agensi

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 15 days ago

    • Free with an Agensi account

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Network Access
    Listed15 days ago

    What's inside

    Frequently Asked Questions

    The skill extracts text, visual figures, and complex tables, converting them into a structured Markdown document that preserves the original teaching flow and LaTeX-formatted formulas.

    Yes, you can specify the depth of content and use the incremental update feature to add new chapters to an existing file without losing your own manual annotations.

    The skill is optimized for Typora but generates standard Markdown and LaTeX, making it fully compatible with Obsidian, VS Code, and other popular personal knowledge management tools.

    Unlike generic prompts that often summarize or skip 'minor' details, this skill uses a structured parsing logic to ensure strict sequence retention and technical accuracy required for university-level study.

    It works best with digital PPT, PPTX, and text-based PDF files; while it can process scanned PDFs, the accuracy of the structural navigation depends on the visual quality of the scan.

    New in Technical Documentation