shared

User-invoked shorthand that inverts delegation — produce a filesystem bundle (checklist, drop-bin directory, notes template) of human-only actions and artifacts blocking project progress, and launch the user's editor on it.

Explicit-request macro. Activate only when the user directly requests this macro; never infer activation from task characteristics. Skill — Shorthand that tells the agent to ask clarifying questions instead of making decisions on its own

Explicit-request macro. Activate only when the user directly requests this macro; never infer activation from task characteristics. Skill — Sequence a task as ordered parts that add up to the whole. Each iteration may read prior iterations' outputs.

Explicit-request macro. Activate only when the user directly requests this macro; never infer activation from task characteristics. Skill — Run N blind agents on the same job in parallel, then merge findings into consensus/unique/conflicts. Agents report only — no edits for concurrent safety.

Explicit-request macro. Activate only when the user directly requests this macro; never infer activation from task characteristics. Skill — Use subagents to save context space and/or parallelize subtasks where possible

Explicit-request macro. Activate only when the user directly requests this macro; never infer activation from task characteristics. Skill — Spawns an independent subagent to critique recent work with web research. Use when the user wants a second opinion, says "doubt this", or wants to verify code against docs.

Explicit-request macro. Activate only when the user directly requests this macro; never infer activation from task characteristics. Skill — Search prior coding-agent chat transcripts (Claude Code, Codex, ...) for context. Use when the user explicitly activates this skill or asks to recall or dig up something from past chats.

Explicit-request macro. Activate only when the user directly requests this macro; never infer activation from task characteristics. Skill — One-shot failsafe — describe what you would do for the current request instead of doing it, then wait for confirmation

Explicit-request macro. Activate only when the user directly requests this macro; never infer activation from task characteristics. Skill — Split a task into parallel chunks, dispatch subagents, and consolidate results

Explicit-request macro. Activate only when the user directly requests this macro; never infer activation from task characteristics. Skill — Scaffold a custom macro — an ordinary SKILL.md that behaves as a first-class member of the macros ecosystem. Writes to project scope (this repo) or user scope (all your projects) and pre-fills the composition line so the new macro composes with other skills.

Explicit-request macro. Activate only when the user directly requests this macro; never infer activation from task characteristics. Skill — Activate Orchestrator Mode — delegate execution to subagents by default and operate at the high level, conserving your own context for direction and synthesis

Explicit-request macro. Activate only when the user directly requests this macro; never infer activation from task characteristics. Skill — Activate Rigor Mode — prioritize correctness, thoroughness, and web-grounded verification over minimalism while active

Explicit-request macro. Activate only when the user directly requests this macro; never infer activation from task characteristics. Skill — Run N serial blind passes on the same job with commits between. Each pass works on the current code state with fresh eyes. Requires a clean worktree.

Explicit-request macro. Activate only when the user directly requests this macro; never infer activation from task characteristics. Skill — Shorthand that tells the agent to avoid time estimates — hours, days, calendar dates, or size buckets that map to time. Describe complexity and scope instead.

Explicit-request macro. Activate only when the user directly requests this macro; never infer activation from task characteristics. Skill — Shorthand that tells the agent to prefix newly created files/folders with a `yymmdd-HHMMSS` timestamp from the `date` CLI. One timestamp per logical job bucket. One turn only.

Explicit-request macro. Activate only when the user directly requests this macro; never infer activation from task characteristics. Skill — Flag content that only makes sense if you were in the room when it was written — defensive justifications, phantom counterarguments, backstory, speculation

Implement an existing specification

Write a phased specification document

Write a specification document

Runs OpenAI Codex CLI to critique a spec or code file. Use when the user wants external AI review, a second opinion, or says "codex critique".

Runs Google Gemini CLI to critique a spec or code file. Use when the user wants external AI review, a second opinion, or says "gemini critique".

Provides battle-tested patterns, best practices, and code examples for building Phaser 3 games. Use when writing game code, implementing game mechanics, setting up scenes, handling physics, animations, input, or any Phaser-related development. Covers architecture, performance, algorithms, and common pitfalls.

Creates timestamped technical analysis and debugging reports following a standardized structure. Use when documenting completed work, analyzing technical issues, investigating bugs, or when the user requests a report.

Evidence-based debugging protocol using the scientific method. Requires one-time project initialization before use. Use when standard debugging has failed and maximum rigor is needed.

Creates effective SKILL.md agent skills following best practices. Use when the user asks to create a skill, write a SKILL.md, or needs help authoring agent instructions.

Conducts Socratic study sessions grounded in a URL or local document. Calibrates to user familiarity, persists session notes to markdown files, and supports resumption across sittings. Use when the user explicitly activates this skill or wants to study a specific document or article.

Author accessible semantic-zoom HTML documents whose structure and content can vary freely across ordered detail levels. Use when creating a zoomable document, converting or adapting existing material into one, or synthesizing sources into a self-contained HTML document with semantic markup, progressive enhancement, and optional per-section detail controls.

Conducts structured interviews to explore any topic through conversational discovery — from software requirements to personal decisions, creative projects, or life planning. Use when the user explicitly activates this skill or needs to think through something via guided conversation.

Run a fully guided, artifact-backed prompt optimization workflow. Use when the user wants an application prompt, prompt builder, agent instruction, routing prompt, or LLM workflow improved against user-provided train/val cases without directly editing source files.