# Cavecrew > Decision guide for delegating to caveman-style subagents. Tells the main thread WHEN to spawn `cavecrew-investigator` (locate code), `cavecrew-builder` (1-2 file edit), or `cavecrew-reviewer` (diff review) instead of doing the work inline or using vanilla `Explore`. Subagent output is caveman-compressed so the tool-result injected back into main context is ~60% smaller — main context lasts longer across long sessions. Trigger: "delegate to subagent", "use cavecrew", "spawn investigator/builder/reviewer", "save context", "compressed agent output". Source: https://skillsagentes.com/skills/juliusbrussee/caveman/cavecrew Repository: https://github.com/JuliusBrussee/caveman Author: juliusbrussee License: MIT Updated: hace 2 meses Context cost: 139 tok installed, 984 tok once triggered, 1.7k tok with every bundled file Bundle: 2 files, 7 KB Permissions requested: none declared ## Install ```bash npx -y skills add JuliusBrussee/caveman --skill cavecrew --agent claude-code ``` ## What it does - Guía de decisión para delegar a subagentes estilo caveman: cuándo lanzar cavecrew-investigator (localizar código), cavecrew-builder (editar 1-2 archivos) o cavecrew-reviewer (revisar un diff) en vez de hacerlo inline. - La salida de estos subagentes viene comprimida en estilo caveman, así el resultado inyectado de vuelta al contexto principal es ~60% más pequeño. - Define contratos de salida exactos por agente: formato path:line, símbolos entre backticks, tokens terminales como "too-big." o "needs-confirm.". - Describe patrones de encadenamiento: localizar → arreglar → verificar, scouting en paralelo, o edición directa cuando el sitio ya se conoce. ## Use it when - Decidir si delegar una tarea a un subagente en vez de hacerla inline. - Localizar código, hacer una edición quirúrgica de 1-2 archivos, o revisar un diff/rama/archivo. - Sesiones largas donde conservar contexto principal importa (resultado comprimido). ## Don't bother when - cavecrew-builder si el archivo aún no se conoce: primero investigator, o se gastan tokens innecesarios. - Refactor de 3+ archivos o cross-cutting: usa el hilo principal o un code-architect, cavecrew-builder devuelve "too-big.". - Pedir feedback general a cavecrew-reviewer: solo devuelve hallazgos, no opiniones de arquitectura. ## What triggers it - "usa cavecrew para encontrar dónde se define X" - "delega esta edición de 2 archivos a cavecrew-builder" - "revisa este diff con cavecrew-reviewer" ## Before you install - Ninguno explícito: es una guía de decisión sobre cuándo invocar los subagentes cavecrew-investigator/builder/reviewer ya empaquetados en el repo. ## Files - README.md — 3 KB - SKILL.md — 4 KB ## SKILL.md Reproduced verbatim from JuliusBrussee/caveman under MIT. This section is the upstream document and is in English. Cavecrew = three subagent presets that emit caveman output. Same job as Anthropic defaults (`Explore`, edit-style agents, reviewer); difference is the tool-result they return is compressed, so main context shrinks per delegation. ## When to use cavecrew vs alternatives | Task | Use | |---|---| | "Where is X defined / what calls Y / list uses of Z" | `cavecrew-investigator` | | Same but you also want suggestions/architecture commentary | `Explore` (vanilla) | | Surgical edit, ≤2 files, scope obvious | `cavecrew-builder` | | New feature / 3+ files / cross-cutting refactor | Main thread or `feature-dev:code-architect` | | Review diff, branch, or file for bugs | `cavecrew-reviewer` | | Deep code review with rationale + alternatives | `Code Reviewer` (vanilla) | | One-line answer you already know | Main thread, no subagent | Rule of thumb: **if you'd want the subagent's output in 1/3 the tokens, pick cavecrew. If you'd want prose, pick vanilla.** ## Why this exists (the real win) Subagent tool results get injected into main context verbatim. A vanilla `Explore` that returns 2k tokens of prose costs 2k tokens of main-context budget every time. The same finding from `cavecrew-investigator` returns ~700 tokens. Across 20 delegations in one session that's the difference between context exhaustion and finishing the task. ## Output contracts What main thread can rely on per agent: **`cavecrew-investigator`** ```
: - path:line — `symbol` — short note totals: . ``` Or `No match.` Always file-path-first, line-number-attached, backticked symbols. Safe to grep with `path:\d+`. **`cavecrew-builder`** ``` — . verified: . ``` Or one of: `too-big.` / `needs-confirm.` / `ambiguous.` / `regressed.` (terminal first token). **`cavecrew-reviewer`** ``` path:line: : . . totals: N🔴 N🟡 N🔵 N❓ ``` Or `No issues.` Findings sorted file → line ascending. ## Chaining patterns **Locate → fix → verify** (most common): 1. `cavecrew-investigator` returns site list. 2. Main thread picks 1-2 sites, hands paths to `cavecrew-builder`. 3. `cavecrew-reviewer` audits the diff. **Parallel scout** (when investigation is broad): Spawn 2-3 `cavecrew-investigator` calls in one message (different angles: defs vs callers vs tests). Aggregate in main thread. **Single-shot edit** (when site is already known): Skip investigator. Hand exact path:line to `cavecrew-builder` directly. ## What NOT to do - Don't use `cavecrew-builder` when you don't already know the file. Spawn investigator first or main thread will eat tokens passing context. - Don't chain `cavecrew-investigator → cavecrew-builder` for a 5-file refactor. Builder will return `too-big.` and you'll have wasted a turn. - Don't ask `cavecrew-reviewer` for "general feedback" — it returns findings only, no architecture opinions. Use `Code Reviewer` for that. - Don't expect prose. Cavecrew output is structured, sometimes terse to the point of cryptic. If a human will read it directly, paraphrase. ## Auto-clarity (inherited) Subagents drop caveman → normal English for security warnings, irreversible-action confirmations, and any output where fragment ambiguity could be misread. Resume caveman after. --- Skills Agentes — https://skillsagentes.com/skills/juliusbrussee/caveman/cavecrew