Valida uma análise antes de compartilhar — metodologia, precisão e verificações de viés. Use quando revisar uma análise antes de uma apresentação para stakeholders, verificar cálculos e lógica de agregação, conferir se os resultados de uma query SQL parecem corretos, ou avaliar se as conclusões são de fato suportadas…
Escreve SQL otimizado para o stack Evolution (PostgreSQL primário) com boas práticas. Use quando precisar traduzir uma necessidade de dados em SQL, construir uma query com múltiplas CTEs, joins e agregações, otimizar uma query contra tabelas grandes, ou obter sintaxe específica para consultas no banco do Evo CRM, Evo…
Query MongoDB databases configured in .env (DBMONGON). Use when the user asks to query, explore, or audit data in a MongoDB database. Picks connection by label (e.g. 'orders-dev', 'analytics-prod') or numeric index. Read-only by default — writes refused unless DBMONGONALLOWWRITE=true on that block.
Query MySQL / MariaDB databases configured in .env (DBMYSQLN). Use when the user asks to query, explore, or audit data in a MySQL database. Picks connection by label (e.g. 'orders-dev', 'analytics-prod') or numeric index. Read-only by default — writes refused unless DBMYSQLNALLOWWRITE=true on that block.
Query PostgreSQL databases configured in .env (DBPOSTGRESN). Use when the user asks to query, explore, or audit data in a Postgres database. Picks connection by label (e.g. 'msgops-dev', 'bms-prod') or numeric index. Read-only by default — writes refused unless DBPOSTGRESNALLOWWRITE=true on that block.
Inspect Redis instances configured in .env (DBREDISN). Use when the user asks to read keys, check cache state, list keys by pattern, or query info/dbsize on a Redis server. Picks connection by label (e.g. 'cache-dev', 'queue-prod') or numeric index. Read-only in v1 — no SET/DEL/FLUSH exposed.
Post-AI-generation cleanup — remove single-use helpers, over-abstractions, generic naming, and unnecessary configurability. Use after dev-autopilot or any LLM-generated code burst.
Advisory router — query Claude, Codex, or Gemini for a quick second opinion. Experimental — only Claude is guaranteed available; other models require dev-mcp-setup configuration.
Full autonomous execution from idea to working code. Use when user says "autopilot", "autonomous", "build me", "create me", "make me", "full auto" — orchestrates spec → plan → code → QA → validation across the engineering layer agents.
Cleanly stop an active engineering layer flow (dev-autopilot, dev-deep-interview, dev-plan) and report what was completed. Use when the user says "cancel", "stop", "abort", "cancelomc", or wants to exit a multi-phase dev workflow.
Tri-model orchestration — run the same task through Claude + Codex + Gemini in parallel and synthesize the best answer. Use for high-stakes decisions where multi-model consensus reduces single-model bias.
Two-stage investigation — causal trace (via @trail-tracer) followed by requirements crystallization (via @echo-analyst + dev-deep-interview). Use when a problem is both unclear in cause AND unclear in scope.
Socratic questioning to crystallize vague requirements into a testable spec. Use when input is ambiguous, when the user asks for "deep interview", or before dev-autopilot if the brief is too broad. Pairs with @echo-analyst.
Deep codebase initialization — generate hierarchical AGENTS.md / CLAUDE.md context files for a new project so engineering agents have full context from session start.
Parallel external documentation lookup. Spawn multiple @scroll-docs agents simultaneously to fetch SDK refs, API docs, and framework guides for different topics in one pass.
Meta-skill — extract reusable skills from the current conversation. Identifies patterns that appeared 3+ times and proposes them as new skills. Pairs with skill-creator builtin.
Strategic planning with optional interview workflow. Use when the user wants to plan before implementing — "plan this", "let's plan", structured requirements gathering for vague ideas. Pairs with @compass-planner agent. Supports consensus mode and review mode.
Worktree-first development environment — create per-issue or per-PR git worktrees, optionally with tmux sessions, so multiple work streams don't collide.
Persistence loop — keep working on a task until it's resolved or until a circuit breaker stops you. Different from /loop (which is time-based) — ralph is iteration-based with state.
Multi-perspective consensus planning with Planner/Architect/Critic loop. Use when high-stakes decisions need RALPLAN-DR structured deliberation (auth, migrations, public APIs, irreversible changes). Pairs with @compass-planner + @apex-architect + @raven-critic.
Release preparation — changelog generation, version bump, tag creation. Generic version (not project-specific). For EvoNexus releases, use custom-release instead.
Persist context across engineering sessions — save important decisions, gotchas, or patterns that should survive beyond the current conversation. For long-term workspace memory, prefer mempalace.
★not rated 548▲
+2 5mo agoA38 tokens
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: