Agent skills for Grok Build

Test driven development. Use when implementing a feature or bugfix. Write the failing test first, then the code that passes it.

Confirm the work is actually done. Use when you ask whether something is finished. Requires evidence: a command that ran, a page that…

5 Whys, Fishbone diagrams, Fault Tree Analysis, and systematic debugging approaches.

Diagnose software failures through reproduction, boundary localization, competing hypotheses, and discriminating experiments.

Hunting skill for business logic vulnerabilities.

Generate personalized status briefings on demand.

Create, update, or delete Route 53 DNS records (project subdomains) using least-privilege AWS credentials from Azure Key Vault.

Automate Route4me tasks via Rube MCP (Composio).

Keep routes clean and focused on mapping requests to controllers; avoid business logic, validation, or database operations in route files

This skill should be used when the user asks a broad engineering question that could match several areas, asks "which skill covers this"…

Builds routing and origin-destination analysis workflows in CARTO.

Use when choosing reviewers, domain experts, or likely owners for a file or directory from CodeScene project data.

Choose and configure JSON-RPC transports in Nethereum.

Build and audit Arabic/RTL web UI with logical CSS, bidirectional text handling, Arabic typography, RTL-safe component patterns, and icon…

Provides thread-aware debugging and health monitoring for FreeRTOS, RT-Thread, and Zephyr embedded systems.

Review and refactor Ruby code while preserving behavior and keeping the design simple.

When to activate: RSpec, describe context it, let and let!, FactoryBot, factories vs fixtures, request specs, shared examples, verified…

Diagnose Ruflo health, then report system, MCP server, and active-agent status without changing the installation

Design a RULER emotional literacy sequence for recognising, understanding, labelling, expressing, and regulating emotions.

Project-rules enforcer. Reviews recent changes strictly against the binding instructions in CLAUDE.md (and any project rule checklist)…

Execute a bounded Coval evaluation on explicit test cases, then audit conversation, recording and metric evidence.