Explore skills

Safe, reliable file editing with validation and debugging, ideal for code, configs, or markup (HTML, JSON, Python, JS). It prevents common mistakes like wrong parameter names or unintended changes, incorporating validation, search verification, and rollback strategies.

JUnit 5 skill pack for creating, debugging, modernizing, documenting, and routing JVM test workflows.

Inspects a codebase, detects the test stack, and plans or writes isolated unit and component tests safely.

A 4-pass prompt optimization engine that combines 7 skills into a pipeline, guaranteeing 5/5 across all quality dimensions. It takes a user prompt and returns an optimized version with a full scoring report.

Prevent Terraform/OpenTofu hallucinations by diagnosing and fixing failure modes: identity churn, secret exposure, blast-radius mistakes, CI drift, and compliance gate gaps. Use when generating, reviewing, refactoring, or migrating IaC and when building delivery/testing pipelines.

Optimizes Claude's coding collaboration by countering LLM weaknesses and leveraging its strengths like tenacity and iteration. Use it for any coding task, code review, architecture discussion, or software development conversation, triggering on code requests, debugging, or refactoring.

Analyzes GitHub repositories, tools, or AI agent frameworks to extract their best capabilities into native OpenClaw skills, perfect for replicating features or creating new skills from external projects.

This skill safely smoke-tests a Vercel Edge Middleware PR before merge by toggling Deployment Protection, running anonymous curl across multiple Host headers to verify SEO metadata, and then restoring protection. It generates a PR comment with pass/fail and a merge recommendation, ideal for PRs affecting Vercel Edge Middleware or custom domains.

This skill removes common AI-slop behaviors like sycophancy, preambles, and excessive hedging from responses. It applies to all user-facing assistance tasks, from coding to general chat, to reduce user frustration.

A 4-phase root cause investigation for any technical issue, such as bugs, test failures, or unexpected behavior. It enforces finding the root cause before attempting fixes, disallowing random solutions.

Analyzes the current Claude Code session (conversation history and changed code files) to detect potential bugs, technical debt, regressions, and performance issues, then creates structured GitHub issues for each one after user approval.

Analyzes source code logic, generates Mermaid flowcharts or SVG architecture diagrams, and renders them as PNG images. Supports 15 themes, 5 visual styles, a semantic graphic system, and 40+ product icons.