Code Review skills

This skill helps you query and manage Greptile-indexed repositories via its REST API to answer codebase questions efficiently.

This skill generates LLM-optimized project context from a codebase directory, delivering a structured markdown snapshot for rapid comprehension.

This skill guides you to write concise CLAUDE.md files that maximize Claude code performance while preventing instruction overload.

This skill orchestrates multi-model roundtables to design optimal workflows and execute multi-perspective analyses with grounded synthesis.

This skill helps you avoid common Linux kernel mistakes by providing guidance on atomic context, allocation failures, and locking practices.

This skill helps you select cost-efficient AI models matched to coding or reasoning tasks, optimizing quality and expenses across workflows.

This skill helps perform ethical security reviews, align with OWASP guidance, and generate actionable bug bounty and remediation steps for scoped programs.

This skill helps orchestrate multi-agent collaboration by creating local chat rooms with channels, claims, and durable summaries in the workspace.

This skill enforces architecture-first delegation for AI coding agents, ensuring every line of code fits the system before it's written.

This skill helps AI agents reduce hallucinations by running adversarial debates between agent and anti-agent, iterating until validation.

This skill executes task plans by dispatching independent subagents with two-stage reviews for spec compliance and code quality, accelerating multi-task

This skill blocks destructive edits to protected identity files, ensuring safe writes and preserving CLAUDE.md and other critical identities.

This skill audits authentication flows for security vulnerabilities, helping you strengthen login, session management, and token handling.

This skill enforces structured subskill organization by placing outputs in data/ and code under subskills/<feature>/, keeping the root clean.

This skill audits Google Analytics implementations for common issues, configuration gaps, and optimization opportunities effectively.

This skill acts as an AI pair programmer to debug code, explain algorithms, and review PRs, boosting engineering productivity.

This skill audits HTML/CSS for WCAG 2.1 compliance, highlighting color contrast, ARIA labels, keyboard navigation and screen reader support.

This skill generates structured .code-task.md files from descriptions or PDD plans, auto-detecting input type and producing clear Given-When-Then acceptance

This skill securely exposes your local workspace through a temporary code-server tunnel for your owner to view, browse, and inspect code.

This skill analyzes code for security risks, detects dangerous patterns, and provides actionable guidance to reduce exposure before installing untrusted skills.

This skill helps you ensure your skills pass ClawHub security scans with benign ratings by enforcing a JSON metadata format and strict security checks.