project-twelve

Run ProjectTwelve Unity batch validation and EditMode/PlayMode test suites via Unity 6000.5.1f1. Use when the user or ticket asks to run Unity tests, EditMode fixtures, batchmode validation, or verify test results from TestResults/*.xml.

Log agent sessions, consolidate project memory (dream), promote reviewed memory, and load context for new chats. Use when ending a session, improving cross-session recall, or when the user mentions agent memory, dream, log-session, or memory.md.

Wiki tickets under docs/wiki/tickets/ are the canonical task queue, each linked to a GitHub issue. Use when picking work, claiming an issue, or preparing PR references.

Recommend the next open wiki ticket to work on, ranked by phase then priority, without claiming or mutating anything. Use when the user asks "what's next", to pick the next task, or to start work without a specific ticket in hand.

Sync project-twelve with the Assets/_Licensed submodule via fetch_remotes.py, git hooks, and /fetch-remotes. Use at session start, before picking backlog work, or when the user asks to sync repos or get latest.

Use when the user asks to commit and push, publish, ship, or prepare a release commit. Guides staging only intended changes, writing a Conventional Commit, committing, and pushing to origin after verifying status and diff.

Apply Karpathy's anti-pattern checklist before/during non-trivial coding — surface assumptions, resist overcomplication, keep diffs surgical, frame work as verifiable goals. Use on ambiguous or multi-step tasks, or when the user asks for a "Karpathy pass".

Use when creating, reviewing, or improving first-party agent skills in this repo. Apply whenever a user asks to add a skill, edit SKILL.md, optimize skill triggering, benchmark skill behavior, or adapt patterns from Anthropic's public skills into the canonical .claude skill tree.

Use when encountering a bug, failing test, build failure, unexpected behavior, flaky behavior, regression, performance anomaly, or integration issue. Apply before proposing fixes so the agent finds root cause, compares working patterns, tests one hypothesis at a time, and avoids guess-and-check patches. Also use when the user asks to diagnose, investigate, root-cause, or fix a flaky failure.

Use when implementing a feature, bug fix, refactor, or behavior change that can be tested. Apply before production code changes to enforce red-green-refactor, prove tests fail for the intended reason, and keep implementation minimal. Also use for risky vertical slices or when the user asks for TDD, red/green/refactor, or tests first.

Use before claiming work is complete, fixed, passing, ready to commit, or ready for PR. Apply to ensure fresh command output supports every success claim and to report warnings or failures honestly.

Generate and manage pixel-art game assets through PixelLab's MCP server, including characters, animations, sidescroller and top-down tilesets, isometric tiles, map objects, UI panels, and fonts. Use when the user asks for PixelLab/Vibe Coding, pixel-art sprites or animations, game tilesets, HUD art, or other PixelLab-generated assets; also use to poll, download, review, list, or delete PixelLab jobs and assets.

Compare MCP code intelligence tiers — CLI wrappers, ast-grep, Serena, Zoekt, fff file search, embeddings. Prefer text search before heavy indexes.

Commit and push changes in Assets/_Licensed (project-twelve-assets), then bump the submodule gitlink in project-twelve. Use after editing licensed docs, catalogs, or art; pairs with repo-sync for fetch/pull.

Router for CLI tooling skills — install checklist, safe patterns, search, git, MCP. Start here for agent command-line workflows.

Deep bug-finding on recent commits: high-severity correctness only (data loss, crashes, security, major breakage). Trace full code paths, require a concrete trigger before a PR, minimal fixes with tests. Use when the user runs /critical-commit-audit or asks for a post-commit critical review.

Query JSON/YAML/config with jq and yq; HTTP probes with curl; local sqlite inspection. Use for structured data, not raw grep on minified JSON.

Capture task quality signals and log them to agent memory to build a feedback loop. Use at the end of each implemented task or merged PR, or whenever a task required more than one agent iteration.

Safe git and GitHub CLI workflows — status, diff, log, branch, PR operations with inspect-before-commit patterns.

Human workstation provisioning for agent CLI tools via winget, apt, and Homebrew. Not for automated agent install loops — user runs these once.

Run linters and security scanners with bounded scope — ruff, prettier, eslint, shellcheck, trivy, hadolint.

Design and implement an MCP server's tools, resources, and prompts with safe transport and input validation. Use when building or extending a Model Context Protocol server for agent integration.

Bump the project version with a semver rubric, promote the Unreleased changelog section, and verify before committing. Use when the user asks to cut a release, bump the version, or tag a new version.

Reusable command rules that minimize accidental data loss, context overflow, and unreviewed changes. Use before editing, searching, and committing.

Fast repo navigation with ripgrep, fd, bat, tree, eza, zoxide, and delta. Use for text search and file discovery — see search-tool-selection for tool choice.

Choose the right search tool before running commands — fd vs rg vs grep vs ast-grep vs fff MCP vs IDE Grep. Use at the start of any code-finding task.

Syntax-aware search with ast-grep and semgrep; ctags for symbol indexes. Use when ripgrep has too many false positives — see search-tool-selection first.

Runs detect_subagents then run_subagent (codex/gemini) in review mode via subagent-orchestrator MCP, reads .agent-runs outputs, and summarizes findings. Use when the user asks for external CLI review, codex/gemini review, subagent panel, tie-breaker, or invokes /check-subagents, /run-codex-review, /run-gemini-review, /run-subagent-review.

Run project tasks, manage language versions, and install developer tools reproducibly. Use after inspecting project entry-point files.

Threat-model agentic tooling — MCP/tool abuse, hook/prompt injection, remote control surfaces, and secret leakage. Use during security reviews or when adding features that expose the system to AI agents or remote callers.

Verify an implementation against its spec's acceptance criteria, marking each AC Verified, Failed, or Unknown with concrete evidence. Use after /implement or /test-and-fix, before /pr-ready, or when the user asks whether the spec is actually satisfied.

Native Windows agent workflows — PowerShell, winget, Cursor IDE, Docker Desktop, when WSL is optional.

WSL2 Ubuntu agent workflows — repos, build/test, ast-grep, MCP servers, CI-like commands.