Build MCP servers — research, implement, test, evaluate. Rust (rmcp, default), TypeScript, or Python. Tool design, strict schemas, pagination, actionable errors, spec 2026-07-28 transports, plus a 10-question evaluation that scores the result. Triggers:…
Terse posture — drop articles/filler/pleasantries/hedging. Keep technical substance, code blocks, errors exact. Mirrors user's language; never announces itself. Not for onboarding, pedagogy, security warnings. Triggers: 'caveman mode', 'modo caveman', 'be…
Lazy-senior-dev posture for code output — before writing, climb the ladder: YAGNI → reuse what's in the codebase → stdlib → native platform → installed dep → one line → minimum that works. Never cuts validation, error handling, security, or accessibility.…
Clean code lifecycle for Go/Rust/TypeScript/Bun/Python — writing, reviewing, refactoring. Naming, functions, DRY, code smells, duplication (literal/logical/structural), dead-code removal with false-positive guardrails, safe refactoring. Triggers: 'clean…
Scaffold new ValarMindSkills per project conventions. Builds SKILL.md, references/, EXAMPLE files. Triggers: 'criar skill', 'nova skill', 'scaffold skill', 'create skill'.
Over-engineering review — finds what to DELETE, not what to fix. One line per finding with tags delete/stdlib/native/yagni/shrink and a net-lines score. Three scopes: current diff (default), whole repo ('repo'/'audit'), or 'debt' to harvest ponytail: comments…
Read-only planning: analyze and design without changing project files. Writes one step-by-step plan at project root (default IMPLEMENTATION_PLAN.md). Composable with other skills (e.g. /code-security-review /only-plan). Triggers: /only-plan, 'only plan',…
Lifecycle code review Go/Rust/TS/Python. Auto-detects toolchain, runs static analysis, emits severity-ranked findings + file:line evidence, diffs, risk tags (SAFE/REVIEW/BREAKING). Covers OWASP Top 10, perf anti-patterns, test quality. Read-only — every…