world-of-claudecraft

Audit World of ClaudeCraft Codex support for current conventions, model neutrality, instruction quality, agent safety, skill design, hooks, CI integration, and documentation drift. Use when reviewing or modernizing AGENTS.md, .codex, .agents, Codex workflows, Codex-facing scripts, or the repository AI architecture.

Implement World of ClaudeCraft features, bug fixes, and focused refactors using module-first design and behavior-driven tests. Use when adding behavior, fixing a defect, extracting logic from a large file, or moving code behind an existing architectural seam.

Plan a large or multi-session World of ClaudeCraft feature as architecture-aligned vertical slices with explicit decisions, verification, and handoff state. Use when the user requests a phased implementation plan, planning packet, session breakdown, or scalable approach before implementation.

Draft and create a focused GitHub issue for World of ClaudeCraft using the project issue format. Use only when the user explicitly asks to file, create, or open an issue. A request for a draft does not authorize creating the issue.

Run end-of-contribution QA for World of ClaudeCraft, including readiness checks, scoped regression testing, and conditional specialist review. Use when asked to QA changes, make work ready, verify a branch or worktree, or assess whether a contribution is complete.

Perform a whole-tree World of ClaudeCraft release malware audit using the deterministic scanner plus contextual triage. Use before a release, for a release tag or branch, after suspicious dependency or install changes, or when explicitly asked for a malware-focused security review.

Audit a release merge into a long-lived World of ClaudeCraft branch for semantic conflict damage, legacy drift, missed inventory updates, and invalidated planning assumptions. Use after merging a release branch, when reviewing a release merge commit, or before resuming feature work after upstream integration.

Review a World of ClaudeCraft pull request against its actual diff, canonical architecture, tests, and domain risks. Use when asked to inspect, review, assess, or prepare feedback for a pull request. Reading and drafting are allowed by default, but posting GitHub feedback requires explicit authorization.

Audit a release/** merge into a long-lived feature branch for the known drift hazards. Use immediately after merging a release branch into a feature branch (or when reviewing such a merge commit): finds branch-owned files the merge touched, release changes landed on a legacy arm that a migrated handler must mirror, new routes missing surface-corpus rows, release-refactored helpers that need re-binding at their injection sites, and planning-doc premises the merge silently invalidated.

Review a GitHub pull request for World of ClaudeCraft the way the maintainer does. Use when asked to review a PR, look at a PR, or check a PR (a github.com/.../pull/<n> link or a PR number). Fetches the PR, classifies the change by domain (sim, wire/parity, render, ui, server, i18n), verifies the repo's invariants against the real code (not the PR description), checks the merge-conflict scope, independently confirms any consequential finding, then posts a short, plain, friendly GitHub review with severity-tagged findings. Not for reviewing the local working tree (that is /code-review).

Create a GitHub issue for World of ClaudeCraft in the maintainer's house format. Use when asked to file, open, or create a GitHub issue or bug report from a description, a screenshot, or a rough placeholder. Rewrites the input into a clean, professional issue (Problem, Steps to reproduce, Expected, Actual, Scope, Acceptance criteria) and includes a Screenshot section only for UI/UX issues. Posts to levy-street/world-of-claudecraft via gh.

Build features and fix bugs the clean, scalable way for World of ClaudeCraft. Use when adding a feature, refactoring logic out of a large file (sim.ts, hud.ts, renderer.ts, main.ts), or fixing a bug, especially when the change would otherwise append a block of new logic to an existing big file. Extracts self-contained behavior into a small, well-named, unit-tested module behind one of this repo's existing seams instead of growing a monolith, and fixes bugs test-first (reproduce with a failing test, then make the smallest change that turns it green). Keeps merge conflicts small and the codebase scalable for many contributors.

Break a big feature into a phased implementation plan with starter prompts, progress tracking, and cross-session state. Opus 4.8 native - each phase runs as its own session with context-saving subagents, agent teams, and (for batch-heavy work) deterministic Workflows, plus a web-research pass for any third-party surface. Use when a feature is too large for one session.

Release-gate scan for deliberately planted malicious code (crypto miners, data/secret exfiltration, backdoors, RCE/obfuscation, install hooks) across the whole working tree of World of ClaudeCraft. Use before tagging or shipping a release, or whenever you want to confirm the tree is free of malicious code. Runs the deterministic scanner, fans the read-only release-malware-audit agent across categories, and returns a single PASS / BLOCK verdict with confirmed findings and dismissed false positives. Distinct from privacy-security-review, which catches accidental security mistakes.