claude-stuff

Create and manage behavior specification wikis in Obsidian format. Use when creating specs, documenting features, or when user mentions "wiki", "spec", "feature", or "Obsidian".

ONLY USE WHEN REPO IS USING AXIOM LOGGING. Audit Axiom logs to identify and prioritize errors and warnings, research probable causes, and flag log smells. Use when user asks to check Axiom logs, analyze production errors, investigate log issues, or audit logging patterns.

This skill should be used when reviewing pull requests, performing comprehensive code review, analyzing code changes before merge, or when the user asks for thorough/ultra-critical code review. Performs EXTREMELY CRITICAL 6-pass analysis identifying runtime failures, code consistency issues, architectural problems, environment compatibility risks, and verification strategies. Posts structured review as GitHub PR comment. Use when user asks to "review PR", "review this code", "review changes", "check this PR", "analyze PR", "post review", or for Phase 3 of devflow. Supports parallel review mode with multiplier (code-review-3, code-review 6X) for consensus-based reviews. This is an ultra-critical reviewer that does not let things slip and desires only perfection.

ONLY USE THIS SKILL UPON USER/HOOMAN REQUEST TO USE EXACTLY THIS SKILL, ONLY RUN ON HOOMAN [OK]. Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes

Guidelines for using skills effectively - load relevant skills before complex tasks, not every message

Upgrade Obsidian wikis to latest format. Handles structure migration (phases/tasks → features/specs), comment format upgrades (adding emoji prefixes), and Johnny Decimal feature organization. Use when user mentions "upgrade wiki", "migrate wiki", "update wiki format", or has old-format comments.

This skill should be used when the user asks to "use delphi", "ask delphi", or wants multiple parallel oracle investigations of the same question to discover divergent insights. Delphi launches multiple oracle agents simultaneously with identical prompts, allowing them to independently explore and potentially discover different paths, clues, and solutions. Results are saved to .oracle/[topic]/ and synthesized into a final document.

This skill should be used when the user asks to "use the oracle" or "ask the oracle" for deep research, analysis, or architectural questions. The oracle excels at multi-source research combining codebase exploration and web searches, then synthesizing findings into actionable answers. Use for complex questions requiring investigation across multiple sources, architectural analysis, refactoring plans, debugging mysteries, and code reviews.

This skill should be used when analyzing video files. Claude cannot process video directly, so this skill extracts frames hierarchically - starting with a quick overview, then zooming into regions of interest with higher resolution and temporal density. Use when asked to watch, analyze, review, or understand video content.

You MUST use this, UNLESS the human says otherwise, before any EXTENDED creative work (not simple execution) - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

Use when starting a session on a project, returning after time away, or before making significant changes. Essential for building comprehensive understanding of project state through total recall and deep exploration.

Apply Strunk's timeless writing rules to ANY prose humans will read—documentation, commit messages, error messages, explanations, reports, or UI text. Makes your writing clearer, stronger, and more professional.

Create a PR for this branch. ALWAYS links related issues and uses closing keywords to auto-close them on merge.

Coordinator workflow for orchestrating dockeragents through fix-review-iterate-present loop. Use when delegating any task that produces code changes. Ensures agents achieve 10/10 quality before presenting to human.

This skill should be used when parallelizing multi-issue sprints using git worktrees and parallel Claude agents. Use when tackling multiple GitHub issues simultaneously, when the user mentions "blitz", "parallel sprint", "worktree workflow", or when handling 3+ independent issues that could be worked on concurrently. Orchestrates the full workflow from issue triage through parallel agent delegation to sequential merge.

Create or update comprehensive GitHub issues that capture ALL context, requirements, and implementation details. Use when creating or updating tickets, issues, feature requests, or bug reports. Ensures no context is lost between ticket creation and implementation - the moment you ask for a ticket, there's maximum context available that will be lost if not captured now.

Update all documentation in .plans, CLAUDE.md files, docs, and .tasks to match current codebase state. Use when user asks to update docs, refresh documentation, sync docs with code, or validate documentation accuracy.

Extract comprehensive JSON design specifications from visual sources including Figma exports, UI mockups, screenshots, or live website captures. Produces W3C DTCG-compliant output with component trees, suitable for code generation, design documentation, and developer handoff.

Audit implementation progress against a plan, verify completed work, identify remaining tasks, and validate quality. Use when user asks to check plan status, verify implementation, see what's left to do, or validate plan completion.

Create comprehensive implementation plan in .plans directory based on analysis or report. Use when user asks to create a plan, plan implementation, design a solution, or structure work for a feature/refactor/fix.

Code review of current git changes, compare to related plan if exists, identify bad engineering, over-engineering, or suboptimal solutions. Use when user asks to review changes, check git diff, validate implementation quality, or assess code changes.

This skill should be used when creating custom subagents for Claude Code, configuring specialized AI assistants, or when the user asks about agent creation, agent configuration, or delegating tasks to subagents. Covers both file-based agents and Task tool invocation.

This skill should be used when engineering decisions are being made during code implementation. The Archivist enforces decision documentation as a standard practice, ensuring every engineering choice includes rationale and integrates with Architecture Decision Records (ADRs). Use when writing code that involves choosing between alternatives, selecting technologies, designing architectures, or making trade-offs.

MANDATORY for creating Zustand stores. This skill is required when users request state management, creating stores, or mention Zustand. Do NOT create Zustand stores without this skill - all stores must use the required StoreBuilder pattern with immer middleware and factory pattern separation

Control Chrome browser programmatically using chrome-devtools-mcp. Use when user asks to automate Chrome, debug web pages, take screenshots, evaluate JavaScript, inspect network requests, or interact with browser DevTools. Also use when asked about browser automation, web scraping, or testing websites.

Create a new skill that uses an MCP server, following best practices from the MCP CLI guide. Use when user wants to create a skill for a new MCP server or integrate MCP functionality into a skill.

Run comprehensive codebase audit for gaps, deprecated code, TODOs, FIXMEs, architectural anti-patterns, type issues, and code smells. Use when user asks to audit code, find issues, check code quality, or identify architectural problems.

This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.

Audit all filename and naming conventions in the codebase against CLAUDE.md standards and common patterns. Use when user asks to check naming conventions, audit filenames, find naming inconsistencies, or validate file naming patterns.