- Home
- Skills
- Datamktkorea
- Agent Skills
- Readme Writing
readme-writing_skill
0
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill datamktkorea/agent-skills --skill readme-writing- SKILL.md9.7 KB
Overview
This skill automates writing and optimizing a project's README.md following standard conventions and a strict template. It ensures documentation accuracy by extracting real project data and prevents introducing fictional files or settings. The skill is triggered during README creation, review, or refactor tasks to maintain consistent, high-quality docs.
How this skill works
The skill inspects repository artifacts in priority order: dependency manifests (pyproject.toml, package.json, Cargo.toml, go.mod), lock files, infrastructure definitions (Dockerfile, docker-compose.yml, k8s/), environment examples (.env.example, config/), entrypoints (main.py, index.ts, main.go) and project tree. It fills template sections with only verifiable items, generates architecture diagrams when multi-service or external dependencies exist, and omits optional sections unless corresponding files or conditions are found. It validates placeholders against actual files to avoid fictional content.
When to use it
- Creating a new README for a repository
- Updating README after dependency, config, or entrypoint changes
- Refactoring docs to meet organization or open-source standards
- Preparing documentation for release or handoff to other teams
- Automating README generation in CI when manifest files change
Best practices
- Always source values from repository files rather than guessing
- Include only sections that correspond to detected files or settings
- Keep prose concise and fact-based; avoid marketing language
- Use tested commands in Quick Start and verify environment variable names exactly
- Add Architecture only when service boundaries or external dependencies exist
- Validate generated README against the file tree and manifest contents before committing
Example use cases
- Generate README for a Python project using pyproject.toml and main.py
- Update Tech Stack and Quick Start after dependency version changes
- Create Directory Structure and Configuration sections from detected config/ and .env.example
- Produce an Architecture mermaid diagram when docker-compose.yml or k8s manifests are present
- Convert existing loose notes into a standards-compliant README for open-source release
FAQ
No. The skill never invents files, folders, or environment variables. Every placeholder must be filled from existing repository artifacts.
When is the Architecture section included?
Architecture is generated only when the repository shows inter-service communication or external dependencies, such as docker-compose, Kubernetes manifests, or multiple distinct services.