AI-assisted changes can introduce null safety, async, security, and resource bugs before review. This skill gives Claude, Codex, and Claude Code a concise workflow for running UBS as a quality gate.

Teams often add heavy libraries before a React app proves it needs them. This skill guides targeted TanStack adoption after vanilla code works.

Large AI coding projects fail when implementation starts before the architecture is clear. This skill provides a planning workflow for detailed markdown plans, model review, and task handoff.

Developers often need quick Cursor commands without searching external documentation. This skill provides concise CLI guidance for files, folders, diffs, windows, extensions, and remote setup.

AI coding agents lose useful lessons when sessions end. CM turns past work into reusable rules, warnings, and context for Claude, Codex, and Claude Code.

Large markdown plans are hard for agent teams to execute consistently. This skill turns plans into detailed Beads tasks with dependencies, review loops, and handoff guidance.

Parallel AI coding can become hard to coordinate across tasks, files, and reviews. This skill provides prompts and tool steps for NTM, Agent Mail, Beads, and BV.

Multi-agent software projects can stall when agents have fixed roles and fragile handoffs. This skill teaches a shared-task workflow where any Claude, Codex, or Claude Code agent can take the next ready bead.

Manual test case writing is slow and inconsistent across teams. This skill converts requirements into structured Excel-ready test cases with defined fields and coverage guidance.

Teams often add comments that become stale or repeat obvious code. This skill guides AI assistants to write minimal comments that explain non-obvious decisions.

Backend query code often becomes slow, unsafe, or hard to maintain. This skill guides Claude, Codex, and Claude Code toward safer parameters, eager loading, indexes, caching, and transactions.

Backend teams need consistent model definitions that match database and ORM standards. This skill guides AI assistants through model naming, relationships, casts, validation, factories, and seeders.

Database schema changes can break deployments when migrations lack rollback plans or compatibility checks. This skill guides Claude, Codex, and Claude Code toward focused migrations.

Backend API work often stalls on routing, versioning, and response standards. This skill guides Claude, Codex, and Claude Code through consistent REST endpoint design and implementation.

Database teams often need Flyway migrations rewritten for Online Schema Change workflows. This skill converts ALTER TABLE and CREATE INDEX statements into tab-separated OSC lines.

Software issue reports often lack enough evidence for a clear fix. This skill guides Claude, Codex, or Claude Code through staged root cause analysis.

Frontend debugging often needs a real browser session that survives across commands. This skill provides Playwright-based browser control for navigation, inspection, screenshots, and logs.

Repetitive collection transformations and helper logic can obscure Go application code. This skill shows concise samber/lo patterns for collections, errors, concurrency, pointers, and related utilities.

GitHub review and release work can require several commands and repeated context gathering. This skill collects PR context and guides issue or PR changes with structured workflows.

DBX Studio needs consistent assistant behavior across streaming and tool providers. This skill guides prompt edits for tone, tool order, response format, and SQL safety.