Sharper-Flow

Architecture inconsistency detection via deterministic tools, research fallback, and AI heuristic

AI-generated code quality detection via AST, regex, and heuristic analysis

Backend technology stack evaluation — choose boring technology scaled by project context. Use when selecting languages, databases, async infra, or API styles for new or existing backends.

Competitive intelligence research methodology for source code and public information comparison

Improvement-discovery methodology for current-state, LBP, and external landscape analysis

Backlog reconciliation, bug priority assignment, and ROADMAP.md regeneration methodology

Codebase reconnaissance skill for mapping structure, hotspots, and risks

Bounded opportunity scouting for ADV discovery and design phases — identifies missed opportunities and leverage points using existing adv-researcher

MCP tool selection guide — use when choosing between Kagi, Context7, gh_grep, Firecrawl, Playwright, or other MCP tools for a task. Provides decision matrix and anti-patterns.

Context-aware tradeoff analysis — researches codebase and domain to draft criteria questions for user prioritization. Use when facing 2+ viable approaches with meaningful tradeoffs.

PREFERRED: Dual-engine code intelligence (semantic search + symbol lookup). Use INSTEAD of built-in Grep/Glob for code exploration, concept search, and finding implementations by intent. Semantic engine: 92% retrieval quality. Symbol engine: exact function/class lookup, no API key needed.

Generic git worktree boundary guidance. For ADV-managed worktrees, defer to Advance's adv-worktree skill and adv_worktree_* tools.

Ultra-compressed commit message generator. Cuts noise from commit messages while preserving intent and reasoning. Conventional Commits format. Subject <=50 chars, body only when "why" isn't obvious. Use when user says "write a commit", "commit message", "generate commit", "/commit", or invokes /caveman-commit. Auto-triggers when staging changes.

Ultra-compressed code review comments. Cuts noise from PR feedback while preserving the actionable signal. Each comment is one line: location, problem, fix. Use when user says "review this PR", "code review", "review the diff", "/review", or invokes /caveman-review. Auto-triggers when reviewing pull requests.

Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.

morph_edit usage guidance - prefer it for large, scattered, or whitespace-sensitive edits in existing files.

PREFERRED: Dual-engine code intelligence (semantic search + symbol lookup). Use INSTEAD of built-in Grep/Glob for code exploration, concept search, and finding implementations by intent. Semantic engine: 92% retrieval quality. Symbol engine: exact function/class lookup, no API key needed.