Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsBlogPro
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges
  • Chrome Extension
  • Skill Manager

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Readme

ASecurity

Assess the codebase and update the README with any new or outdated information

2 stars
0 votes
0 copies
0 views
Added 9/29/2026
devopsgodockerapidatabaseci/cd

Works with

cliapi

Security Analysis

A100/100

Scanned 9/29/2026

$npx -y skills add abnegate/claudes --skill readme --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Readme?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Readme
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/abnegate-readme/badge)](https://www.skillsdirectory.com/skills/abnegate-readme)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
Files
SKILL.md
---
name: readme
description: Assess the codebase and update the README with any new or outdated information
---

# Update README

Assess the current state of the project and update (or create) the README to accurately reflect it.

**DO NOT fabricate. Read actual files, configs, and code before writing anything.**

## Step 1: Assess the Project (Parallel)

Launch **all four agents in a single message** to build a complete picture of the codebase simultaneously. Do NOT run them sequentially.

**Agent 1 — Project identity and current README:**
- Read `README.md` (if it exists) to understand the current documented state
- Read package manifests (`package.json`, `composer.json`, `Cargo.toml`, `build.gradle`, `pyproject.toml`, `go.mod`, `Gemfile`, `*.csproj`, etc.)
- Read `.claude-plugin/plugin.json`, `plugin.json`, or similar plugin/extension manifests
- Determine: project name, description, version, language, framework, license, author
- Output the full content of the existing README (or note that none exists)

**Agent 2 — Structure and features:**
- List top-level directories and key files
- Identify major modules, packages, or components
- Read entry points (`main.*`, `index.*`, `app.*`, `server.*`, `src/`, `cmd/`, etc.)
- Identify APIs, CLI commands, routes, or other public interfaces
- Check for Docker/container files, CI/CD configs, deployment scripts

**Agent 3 — Setup and usage:**
- Check for Makefiles, Dockerfiles, docker-compose, scripts/, bin/
- Identify build commands, test commands, lint commands from package manifests and scripts
- Check for environment variables (`.env.example`, `.env.sample`, config files)
- Check for database migrations, seed files, or other required setup steps
- Read CONTRIBUTING.md, DEVELOPMENT.md, or similar if they exist

**Agent 4 — Dependencies and configuration:**
- Read dependency files (lock files, requirement files)
- Identify key dependencies and their purpose
- Check for config files that need documenting
- Look for pre-requisites (runtime versions, system dependencies, external services)

**Wait for all four agents to complete before proceeding.**

## Step 2: Diff and Update

Synthesize all four agent outputs. Compare the codebase state against the existing README (from Agent 1) and identify:

- **Missing**: features, commands, modules, config, or setup steps that exist but aren't documented
- **Stale**: entries that reference files, commands, APIs, or patterns that no longer exist
- **Inaccurate**: descriptions that don't match the current code behavior
- **Outdated**: version numbers, dependency lists, or instructions that have drifted

Then edit (or create) the README following these principles:

Edit (or write) the README following these principles:

- **Match the existing style.** If the README uses tables, keep tables. If it uses bullet lists, keep bullet lists. Don't impose a new format on an established README.
- **Preserve intentional content.** Don't remove sections the author wrote manually (architecture decisions, contribution guidelines, acknowledgements) unless they're factually wrong.
- **Add what's missing.** New modules, commands, features, config options, setup steps.
- **Remove what's gone.** Entries for deleted files, deprecated features, old instructions.
- **Fix what's wrong.** Stale version numbers, incorrect commands, outdated descriptions.
- **Keep it concise.** The README should help someone get started and understand the project, not document every implementation detail.

For a new README, include at minimum:
1. Project name and one-line description
2. Prerequisites and installation
3. Usage / getting started
4. License (if determinable from manifests)

Only add additional sections if the project has content worth documenting (API reference, configuration, architecture, contributing guidelines, etc.).

## Step 3: Verify (Parallel)

Launch **two agents in a single message** to verify the updated README:

**Agent 1 — Completeness check:**
- Every documented file, command, or feature actually exists in the codebase
- Every significant public interface or module has a mention
- No entries reference files or commands that don't exist on disk

**Agent 2 — Accuracy check:**
- Install/build/test instructions are runnable (check the actual commands exist in manifests and scripts)
- Version numbers match package manifests
- No placeholder text, TODO markers, or guessed descriptions remain

If either agent finds issues, fix them before reporting done.

## Hard Rules

1. **Read before writing** — every description must come from reading actual code or config, not inferring from names
2. **Preserve voice** — match the existing README's tone and style, don't impose your own
3. **No bloat** — don't add sections the project doesn't need. A 20-line README for a small project is fine.
4. **No lies** — if you can't determine what something does after reading it, say so or omit it. Never guess.
5. **Minimal diff** — change only what's wrong or missing. Don't rewrite sections that are already accurate.

Attribution

abnegateabnegate
View sourceSee grades on GitHubMore from abnegate →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Terraform Module Library

Build reusable Terraform modules for AWS, Azure, and GCP infrastructure following infrastructure-as-code best practices. Use when creating infrastructure modules, standardizing cloud provisioning, or implementing reusable IaC components.

401991 votes

sematext-otel

Wire a service's OpenTelemetry output to Sematext Cloud. Walks through region, App-type, instrumentation flow (managed OTLP endpoint vs Sematext Agent), and signal selection (traces/metrics/logs), then produces the exact env-var block and points at a runnable reference example in this repo. Invoke when instrumenting a new app for Sematext.

01 votes

Deployment Patterns

Deployment workflows, CI/CD pipeline patterns, Docker containerization, health checks, rollback strategies, and production readiness checklists for web applications. Use when setting up deployment infrastructure or planning releases.

2699140 votes

Babysit

Watch a pull request or review cycle until it is ready to merge. Use when asked to babysit, monitor, or keep checking PR comments, reviews, and CI until all actionable issues are resolved.

971540 votes

V7 Roster

Interact with the Paperclip control plane API for task coordination and governance. Use when checking assignments, updating issue status, posting comments, delegating work, managing routines, or calling Paperclip API endpoints.

953190 votes
View all in devops →