Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
Muse
Dots+22 moreClaude CodeCursorCodex CLIMuseDots
OpenClaw
Manus
Grok
Grok Bot+18 agentsREADME
Standardize repository front pages with technical accuracy, verifiable code snippets, and reader-first structure.
$3
README
Example session with this skill installed
Analyze this repository and write a Standard tier README. It is a Python CLI tool for resizing images named 'pix-shrink'. It needs Pillow and Python 3.10.
- Read your context and instructions
- Compiled the readme
I have analyzed the repository structure and manifests. I've drafted a Standard tier README that prioritizes the installation of pix-shrink and provides a verifiable quick start example.
README.md
pix-shrink
Resize images from the command line without losing metadata. Useful for batch processing web assets while maintaining EXIF data.
Requirements
- Python 3.10 or later
- Pillow 10.0 or later
Installation
Install the package using pip
pip install pix-shrink
Quick start
To shrink an image to a specific width while maintaining aspect ratio:
pix-shrink input.jpg --width 800 --output small.jpg
Expected output
Reading input.jpg...
Resizing to 800x600
Saved small.jpg (1.2MB -> 145KB)
Usage
Batch processing
Shrink all images in a directory to 50% scale.
pix-shrink ./images/ --scale 0.5 --suffix _thumb
Configuration
| Name | Required | Default | What it does |
|---|---|---|---|
| PIX_SHRINK_QUALITY | No | 85 | Sets the JPEG output quality (1-100) |
Values are read from environment variables.
Where to go next
| Need | Destination |
|---|---|
| Learn more | Documentation site |
| Reproducible bug | Issue tracker |
| Usage question | Discussions |
| Terms of use | LICENSE |
Next steps
- Add the
PIX_SHRINK_QUALITYvariable to your.env.example. - Verify the documentation site URL in the "Where to go next" table.
Connects securely to your tools. The creator never sees your data.
About this skill
Maintain the gateway to your repository with a tool that treats the README as a functional entry point rather than a static document. This skill addresses the problem of fragmented documentation, unclear setup paths, and incomplete project front-pages that drive away potential users. It ensures every visitor can answer five critical questions: what the project does, who it is for, how to run it, how to verify success, and where to go next.
What it does
- Drafts multi-tier READMEs by selecting between minimal, standard, or advanced templates based on project complexity.
- Audits and scores existing files against strict clarity and structure rules to identify missing setup steps or unclear value propositions.
- Extracts facts from manifests like
package.json,pyproject.toml, and.env.exampleto ensure installation steps are technically accurate. - Standardizes prose by enforcing active voice, second-person imperatives, and removing filler words that obscure technical meaning.
- Verifies runnability by checking that every code snippet is self-contained and includes expected output for user comparison.
How it works
- Fact collection through scanning repository manifests, entry points, and CI workflows to identify requirements and commands.
- Tier selection where the agent picks a prose budget and section list based on whether the project is a script, library, or platform.
- Drafting and editing using a reader-first order that prioritizes the "first success" over internal code architecture.
- Validation using a bundled Python checker to enforce sentence length limits and syntax-correct code fences.
Frameworks & tools
This skill works with any repository type including Python, Node.js, Rust, Go, and Docker. It includes a specific check_readme.py script requiring Python 3.8 or later for automated linting.
Why this beats prompting it yourself
Generic prompts often produce "marketing fluff" and invented facts that lead to broken installation steps. This skill follows a rigid methodology that prevents hallucinated commands and forces the inclusion of observable success criteria for every code example.
Use cases
- New project scaffolding to create a standard-compliant front page before the first commit.
- Documentation audits to score existing READMEs and identify why newcomers are struggling to install the tool.
- Plugin documentation for agent skills or IDE extensions that require specific host-loading instructions.
- Monorepo management to build a high-level map that routes users to specific package documentation.
Known limitations
Does not generate API reference pages, long-form tutorials, or CHANGELOG entries. Relies on GitHub-flavored Markdown standards. Requires Python 3.8+ to run the optional automated checker script.
How to use
Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scanned
Verified clean 6 days ago
- Passed all security checks, Safe to install
Requires
Python