ceasarXuu
Community@ceasarXuu
ceasarXuu publishes 16 coding-agent skills covering multi-agent delegation, engineering planning, debugging, frontend refactoring, three.js game development, and repository documentation.
Agent Skills by ceasarXuu
Showing 16 vetted skills indexed across 1 GitHub repositories.
ceasarXuuagent-commander
Delegate bounded read-only coding analysis to local coding agents and continue their sessions.
ceasarXuudev-loop
Audit and optimize software development feedback loops across builds, tests, CI gates, and caches.
ceasarXuuthreejs-game
Design, build, optimize, test, and release maintainable three.js games across runtime, rendering, physics, and QA.
ceasarXuurules-install
Installs one shared global rule text across multiple local coding agents.
ceasarXuudocs-manager
Audit, organize, and validate repository documentation governance and source-of-truth ownership.
ceasarXuuprovider-switch
Install and manage reversible agent-to-model-provider profiles for Codex and Claude Code.
ceasarXuuclear-prd
Clarify product requirements through structured questions and write a Draft or Ready PRD.
ceasarXuustorybook-skills-standard
Designs and audits Storybook stories, mocks, docs, and quality gates for component-driven frontend workflows.
ceasarXuuplan-report
Generates evidence-based completion reports for implementation plans with verified percentages and engineering benefits.
ceasarXuufrontend-refactoring
Migrates legacy frontend surfaces to isolated v2 UI systems with structure-first inventories and safe cutover.
ceasarXuusummary-my-repo
Generates internal repository summary packs with architecture maps and code-backed evidence snippets.
ceasarXuubug-killer
Diagnose complex bugs with evidence-gated multi-path root-cause investigation and case tracking.
ceasarXuuse-good-plan
Writes and reviews executable engineering plans with protected product authority and pre-phase rebase gates.
ceasarXuushow-my-repo
Generate evidence-backed presentation packs from repository documentation and code.
ceasarXuusubagent-vs-review
Runs bounded adversarial reviews with fresh subagents and a deterministic review governor.
ceasarXuuyour-skill-name
Provides a starter template for authoring new AI agent skill units.
Frequently Asked Questions About ceasarXuu
FAQPage SchemaWhat tasks can I accomplish with ceasarXuu's skills?▼
You can delegate coding analysis between agents like Codex, Claude Code, and OpenCode; write PRDs and engineering plans; debug regressions with evidence-gated root-cause analysis; refactor legacy frontend CSS; build three.js games; audit Storybook stories; and generate repository summaries, investor pitches, and plan-completion reports.
Who are these skills designed for?▼
Software engineers, frontend developers, and technical leads who work with coding agents such as Codex, Claude Code, OpenCode, Pi, TRAE CN, or Grok Build, and who need structured planning, adversarial review, debugging discipline, documentation governance, or game development guidance inside those agent environments.
How do the agent-integration skills work in practice?▼
agent-commander delegates bounded read-only tasks to a locally installed external agent and preserves its session ID for continuation. provider-switch manages reversible agent-to-model-provider configurations such as Codex with DeepSeek, while rules-install synchronizes one shared instruction file across all supported locally installed agents.
What prerequisites do these skills require?▼
Most skills run inside a coding agent such as Codex. agent-commander, rules-install, and provider-switch additionally require the relevant external agents (Claude Code, OpenCode, Pi, TRAE CN, Grok Build) to be locally installed; DeepSeek Harness support is discovery-only in v0.2.1.
How do the planning and reporting skills ensure quality?▼
clear-prd confirms product decisions through multi-round questioning before writing a canonical PRD; se-good-plan produces executable engineering plans rebased on actual implementation; plan-report generates evidence-based completion reports with percentages, verification tables, and next actions; subagent-vs-review adds independent adversarial review with scope-drift detection.