JuliusBrussee

When to delegate to `cavecrew-investigator` (locate code), `cavecrew-builder` (1-2 file edit) or `cavecrew-reviewer` (diff review) instead of working inline or using `Explore`. Their output is compressed, so main context lasts longer.

Compress a memory file such as CLAUDE.md or a todo list into caveman format to save input tokens, keeping a readable backup. Trigger: /caveman-compress.

Ultra-compressed communication mode that cuts output tokens while keeping technical accuracy. Levels: lite, full, ultra and the wenyan variants. Use for /caveman, "caveman mode", "talk like caveman", "be brief" or "less tokens".

Compress a memory file such as CLAUDE.md or a todo list into caveman format to save input tokens, keeping a readable backup. Trigger: /caveman-compress.

When to delegate to `cavecrew-investigator` (locate code), `cavecrew-builder` (1-2 file edit) or `cavecrew-reviewer` (diff review) instead of working inline or using `Explore`. Their output is compressed, so main context lasts longer.

Write a Conventional Commits message compressed to intent only. Use for "write a commit", "commit message", /commit or /caveman-commit.

Find and label every LLM workflow in the repository so Caveman Cloud groups spend by workflow instead of one bucket. Use for "discover workflows" or breaking LLM spend down by workflow.

Read-only review of Caveman Cloud evidence: cost, Cave Score, workflows, traces, latency, errors, routing, savings. Use when asked what Caveman found or where LLM spend goes.

Plan-then-execute implementation against SPEC.md. Native single-thread loop, no sub-agents. On test or build failure, auto-invokes the backprop skill before retrying — a failed verification always considers whether a new §V invariant would prevent recurrence.…

Read-only drift detector. Diffs SPEC.md against current code and reports violations grouped by severity. Writes nothing — suggests remedies via the spec or build skills but never invokes them. Triggers when the user asks to check drift, audit the spec, verify…

Optional design-improvement pass for when you have spare usage to drain. Finds the shallowest modules in the code the spec touches, researches a deeper design, and proposes refactors that shrink interfaces and hide decisions — behavior held constant, tests…

Calibrated interrogation of a fuzzy idea before it becomes a spec. Asks one question at a time, recommends an answer, and lands each answer in §G (goal) or §C (constraints) — unknowns parked as `?` items, never guessed. The cheapest place to kill a bad idea…

Gather external knowledge the spec needs and distill it into §R — the durable research log — so build grounds in facts instead of hallucinating library behavior. Each finding cites a source; unsourced claims are flagged, never written as fact. Triggers when a…

Adversarial senior review of the spec before any code is written. Constructs a skeptical reviewer whose authority comes from the codebase, §R research, and live best-practice — then tries to REFUTE the spec, not rubber-stamp it. Every finding cites evidence…

Create, amend, or backprop bugs into SPEC.md at repo root. Sole mutator of the project spec. Triggers when the user asks to write a spec, start a new spec, distill a spec from existing code, add invariants, amend sections (§G, §C, §I, §V, §T, §B), or record a…

Bug → spec protocol. When a bug is found or a test fails, trace the cause, decide whether a new §V invariant would catch recurrence, append to §B. This is the one non-obvious thing SDD does that plan-then-execute doesn't. Triggers on test failure, bug report,…

Example skill loaded from resources_discover

Scaffold a complete cave plugin bundle — generates .cave-plugin/plugin.json manifest and the standard directory structure (commands/, skills/, agents/, themes/, hooks/). Use when a user wants to create, publish, or package a cave plugin for the marketplace.…

How to write and maintain DESIGN.md as the visual specification layer for Cavekit projects. Nine-section format, design tokens, accessibility, integration with kits/plans. Use when defining or revising visual identity, importing a third-party design system,…

Cavekit specification-driven development methodology — the Hunt lifecycle (Draft → Architect → Build → Inspect → Monitor) and how to apply it. Use when starting a Cavekit project, structuring an existing codebase as kits/plans, or routing between sub-skills.

Trace bugs and manual fixes back to kits and prompts; fix at the source so the iteration loop can reproduce the fix autonomously. Six-step revision process plus the single-failure backpropagation protocol. Use when a manual hot-fix has been applied, when…

Validation-first design for Cavekit — every kit requirement must be automatically verifiable. Six-gate validation pipeline, phase gates, merge protocol, completion signals, acceptance criteria patterns. Use when designing acceptance criteria, wiring CI gates,…

Ultra-compressed communication mode (lite / full / ultra) that cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Use when the user requests "caveman mode", "less tokens", "be brief", or when output budget is tight.

De-slop pass for any text. Scans for the statistical fingerprints of AI writing (negative parallelism / "not X but Y", em-dash density, rule-of-three, false ranges, puffery vocabulary, uniform cadence, hedged both-sidesing), rewrites by meaning, then re-scans…

Finds and finishes the last 20% of a built solution — the experiential layer agents skip: what the output actually reads like, which pages or screens should exist, first-run and empty states, opinionated defaults, the golden path. Use when planning any build…

Install a per-turn canary signal (e.g. starting every reply with the user's name and a turn counter) so silent context degradation becomes visible the moment it happens, and run a recovery protocol when the canary trips. Use when the user mentions a "canary",…

Adversarial senior-engineer review for agent-generated plans, designs, and architectures. Treats the current output as junior work, constructs a senior reviewer whose domain expertise comes from live codebase research plus web research of current best…

Authoritative guide for implementing stunning, accessible, performant UI. Synthesizes design engineering philosophy, accessibility standards, animation principles, spatial design, typography, color systems, and component craft into a single actionable…

Run a spec-driven agent loop where coding tasks live as markdown specs that move through inbox → active → archive, get implemented by Claude Code or Codex, and pass a review gate before they count as done. Use when the user mentions "loop factory", a…

Calibrated grilling session for stress-testing a plan, design, idea, or decision. First assesses the user's topic knowledge, confidence, and desired pressure level, then asks one question at a time with recommended answers. Use when user says "grill me",…

Use when orchestrating Loop Factory specs from factory/specs/inbox or factory/specs/active, dispatching work to Codex or Claude agents, generating implementation prompts, or deciding next loop state.

Use when orchestrating Loop Factory specs from factory/specs/inbox or factory/specs/active, dispatching work, generating implementation prompts, or deciding next loop state.

Use when code changes should be scanned and reflected back into living specs, docs, or architecture notes without changing product intent.

Use when reviewing active Loop Factory specs against the working tree, checking verification evidence, deciding ACCEPTED or CHANGES_REQUESTED, and archiving accepted specs.

Use when code changes should be scanned and reflected back into living specs, docs, or architecture notes without changing product intent.

Use when reviewing active Loop Factory specs against the working tree, checking verification evidence, deciding ACCEPTED or CHANGES_REQUESTED, and archiving accepted specs.

Drive web pages through the caveman-browse MCP tools (browser_snapshot, browser_act, browser_eval, browser_recover). Use whenever interacting with a web page, dashboard, or form through caveman-browse — it teaches the snapshot→act→verify loop, query-first…