Developers need a consistent way to verify code changes. This skill gives Claude, Codex, and Claude Code project-specific test commands.

Teams lose lessons from completed AI coding tasks. This skill guides Claude, Codex, and Claude Code to return concise reflection fields after work.

Project structure can be hard to follow during AI-assisted work. This skill guides Claude, Codex, and Claude Code to read architecture, conventions, patterns, and gotchas.

Teams lose time when generated code ignores established repository conventions. This skill helps Claude, Codex, and Claude Code load project pattern guidance before implementation.

Developers need a consistent way to verify code before handoff. This skill guides Claude, Codex, and Claude Code through project linting and typechecking commands.

Parallel implementation can cause agents to overwrite files from other workstreams. This skill defines ownership, read-only access, and blocked workflows for Claude, Codex, and Claude Code.

Large repositories are hard to understand before making changes. This skill gives Claude, Codex, and Claude Code a repeatable exploration workflow.

Subagent workflows can stall when errors, unclear requirements, or contract issues appear. This skill gives retry, context, split, escalation, and abort patterns.

Teams often repeat design debates or miss prior rationale. This skill guides Claude, Codex, and Claude Code to review decision records before new choices.

Teams lose useful lessons when task reflections stay isolated. This skill aggregates repeated patterns into a compact record for future improvement.

Codebases drift when contributors apply different style choices. This skill loads project conventions so Claude, Codex, and Claude Code can follow shared standards.

Parallel coding tasks often drift when interfaces are unclear. This skill guides agents to define, store, implement, verify, and revise shared contracts.

Long AI workflows can lose context after crashes, timeouts, or session resumes. This skill defines checkpoint files, validation steps, and resume logic for safe recovery.

Complex development tasks often need clear handoffs between planning, implementation, and verification agents. This skill gives Claude, Codex, and Claude Code a structured workflow for coordinating subagents.

AI agents can waste time when requirements or preferences are unclear. This skill gives Claude, Codex, and Claude Code a practical pattern for asking focused questions.

Three.js animation setup can be hard to structure across mixers, clips, bones, and morph targets. This skill gives focused patterns for keyframes, GLTF animations, blending, and procedural motion.

Complex projects often contain independent tasks that waste time when handled sequentially. This skill shows how to plan, launch, collect, and synthesize parallel subagent work.

Slow applications can hide bottlenecks across frontend, backend, database, and algorithm layers. This skill guides profiling, targeted fixes, and before-after measurement.

Git work can become inconsistent across branches, commits, and pull requests. This skill gives Claude, Codex, and Claude Code structured Git workflows, templates, and review checklists.

Many teams add tests without a clear plan, which creates gaps and slow suites. This skill guides test type selection, templates, coverage targets, and validation steps.