tech communication

    1

    Standardize technical documentation, RFCs, and bug reports using CoolShell and Stack Overflow methodologies.

    Free

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    tech communication

    Example session with this skill installed

    I'm having a memory leak in my Node.js app when using the 'sharp' library. Help me write a professional Issue for GitHub.

    • Read your context and instructions
    • Compiled the tech communication

    I have drafted a GitHub Issue following the MRE (Minimal Reproducible Example) principle and Stack Overflow's "How to Ask" standards. This draft prioritizes reproducibility and clear environment context to ensure maintainers can diagnose the leak quickly.

    GitHub Issue Draft: Memory Leak in Sharp during Buffer Processing

    Title: Memory Leak: Heap usage increases indefinitely when calling sharp.webp() on high-concurrency Buffer inputs in Node 18.x

    Problem Description

    A memory leak occurs when processing multiple images into WebP format from memory Buffers. While sharp instances are being garbage collected, the underlying native memory or heap allocated during the .webp() transformation is not being fully released under sustained load.

    Environment

    • Node.js Version: 18.16.0
    • Sharp Version: 0.32.1
    • OS: Debian 11 (Docker)

    Minimal Reproducible Example (MRE)

    const sharp = require('sharp');
    
    async function reproduce() {
      // 1MB dummy buffer
      const input = Buffer.alloc(1024 * 1024, 'a'); 
      
      for (let i = 0; i < 10000; i++) {
        await sharp(input)
          .webp()
          .toBuffer();
        
        if (i % 100 === 0) {
          console.log(`Processed ${i}, RSS: ${Math.round(process.memoryUsage().rss / 1024 / 1024)}MB`);
        }
      }
    }
    
    reproduce();
    

    Expected vs. Actual Behavior

    Expected: Memory usage (RSS) should stabilize after initial buffer allocations and stay within a reasonable range (e.g., < 500MB).

    Actual: RSS grows linearly with every iteration, exceeding 2GB within 5 minutes until the process hits the OOM limit.


    Next steps

    1. Replace the dummy buffer in the MRE with your specific image dimensions if the leak is resolution-dependent.
    2. Run the provided script with node --expose-gc to confirm if manual GC affects the outcome.
    3. Attach the console output logs to the GitHub Issue.

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

    About this skill

    Vague documentation and messy bug reports waste engineering time. This skill standardizes how you write technical designs, READMEs, and troubleshooting requests using the industry-proven methodologies of Chen Hao (CoolShell) and Stack Overflow.

    What it does

    • Standardize technical designs by focusing on problem modeling, mechanism deep-dives, and trade-off analysis instead of just listing features.
    • Generate high-quality READMEs that include clear value propositions, architecture diagrams, and copy-pasteable quick-start guides.
    • Formulate MREs (Minimal Reproducible Examples) for bug reports, stripping away irrelevant logic to isolate the root cause.
    • Structure incident post-mortems using Root Cause Analysis (RCA) that leads with the fix and defines long-term prevention strategies.
    • Refine community questions to meet Stack Overflow standards, ensuring higher response rates through self-explanatory titles and specific behavior comparisons.

    How it works

    1. Identify the scenario such as an RFC, an open-source README, a community help request, or a post-mortem report.
    2. Apply the specific model like the "CoolShell" sharing framework for designs or the MRE principle for debugging.
    3. Verify the output components against mandatory sections like Expected vs. Actual behavior or horizontal solution comparisons.

    Frameworks & tools

    This skill implements the CoolShell technical sharing methodology and Stack Overflow's "How to Ask" guidelines. It utilizes Mermaid for architectural diagrams and follows industry-standard RFC and README structures.

    Why this beats prompting it yourself

    Generic prompts often produce "fluff" or incomplete examples that don't help other developers. This skill enforces strict technical communication constraints, such as mandatory trade-off analysis and the elimination of image-based logs, ensuring every output is actionable and professional.

    Use cases

    • Drafting a technical RFC for a new microservice architecture.
    • Writing a README for a internal library to ensure fast team adoption.
    • Creating a minimal reproduction script to report a bug to an upstream maintainer.
    • Writing a Root Cause Analysis report after a production system outage.

    Known limitations

    Does not provide automated code execution to verify MREs. Requires the user to provide the initial technical context or error logs for refinement.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Technical Documentation.

    Does not verify MREs via execution. Requires manual input of raw logs or code. Optimized for Markdown-based documentation and text communication.

    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 buying a skill.

    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 to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 13 days ago

    • Free with an Agensi account

    Software
    PythonGit

    Permissions

    Read Files
    Listed13 days ago

    What's inside

    Frequently Asked Questions

    This skill transforms vague, unstructured technical notes into standardized documents—like RFCs and bug reports—based on the high-quality communication principles found on CoolShell and Stack Overflow. It eliminates 'fluff' and ensures every piece of documentation includes critical engineering elements like trade-off analysis and minimal reproducible examples.

    You will receive the pre-configured AI agent skill, complete with the logic to process RFCs, READMEs, MREs, and post-mortems. It also includes integrated support for generating Mermaid architectural diagrams to visualize your technical designs.

    While the skill is designed to work with most leading LLMs (like GPT-4 and Claude 3), it performs best with models that have strong reasoning capabilities for analyzing complex technical architectures and logic.

    No complex setup is required; once you import the skill into your preferred AI agent platform, you can immediately start by providing raw technical context, error logs, or draft design ideas for refinement.

    This skill cannot automatically execute or test code to confirm a bug; it relies on the technical data and logs you provide to structure a professional, reproducible report for human review.

    Purchasing this skill grants you access to all future refinements, including updates to documentation templates and adjustments to align with evolving community standards on platforms like GitHub and Stack Overflow.

    New in Technical Documentation

    Free