Makes your AI coding agent answer in terse caveman-style prose and compress memory files like CLAUDE.md, cutting token use while keeping technical facts and code exact. Helps when you want shorter replies and cheaper sessions.
@juliusbrusseeA spec-driven development plugin for Claude Code with skills to build against a SPEC.md, detect drift between spec and code, and feed test failures back into the spec. Frozen, and active development has moved to another project.
@juliusbrusseeEight personal agent skills for now: Caveman base, Interface Kit, Grill Me, Loop Factory, Junior to Senior, Deslopify, Context Canary, and The Last 20%. This repo is shaped by eight things: Caveman - 70k-star token compression without technical loss.
@juliusbrusseeBug → spec protocol. When a bug is found or a test fails, trace the cause, decide whether a new §V invariant would catch…
@juliusbrusseePlan-then-execute implementation against SPEC.md. Native single-thread loop, no sub-agents. On test or build failure,…
@juliusbrusseeWhen to delegate to `cavecrew-investigator` (locate code), `cavecrew-builder` (1-2 file edit) or `cavecrew-reviewer` (diff…
@juliusbrusseeUltra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy.…
@juliusbrusseeTerse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman",…
@juliusbrusseeCaveman encoding for SPEC.md and spec-adjacent writes. Loaded by /spec, /build, /check. Cuts tokens ~75% vs prose while staying…
@juliusbrusseeWrite a Conventional Commits message compressed to intent only. Use for "write a commit", "commit message", /commit or…
@juliusbrusseeCompress a memory file such as CLAUDE.md or a todo list into caveman format to save input tokens, keeping a readable backup.…
@juliusbrusseeFind and label every LLM workflow in the repository so Caveman Cloud groups spend by workflow instead of one bucket. Use for…
@juliusbrusseeRead-only review of Caveman Cloud evidence: cost, Cave Score, workflows, traces, latency, errors, routing, savings. Use when…
@juliusbrusseeRead-only repository explorer for cold-start orientation, broad cross-file localization, or when a direct search failed. Skip it…
@juliusbrusseeQuick-reference card for the three caveman skills and their commands. Trigger: /caveman-help or "caveman help".
@juliusbrusseeSurgical 1-2 file edit. Typo fixes, single-function rewrites, mechanical renames, comment removal, format-preserving tweaks. Hard…
@juliusbrusseeRead-only code locator. Returns file:line table for "where is X defined", "what calls Y", "list all uses of Z", "map this…
@juliusbrusseeDiff/branch/file reviewer. One line per finding, severity-tagged, no praise, no scope creep. Output format `path:line: <emoji>…
@juliusbrusseeGenerate a terse caveman-style commit message for staged changes
@juliusbrusseeCompress a markdown/text file into caveman format to save tokens
@juliusbrusseeQuick reference card for caveman modes, slash commands, and triggers
@juliusbrusseeDrop the always-on caveman activation rule into the current repo for every IDE agent
@juliusbrusseeCaveman-style code review — one-line findings with severity
@juliusbrusseeShow available session usage without estimating savings
@juliusbrusseeActivate megacave mode (caveman in Classical Chinese)
@juliusbrusseeActivate ultracave mode (caveman at maximum compression)
@juliusbrusseePlan-then-execute against SPEC.md. Native Claude Code loop, no sub-agents.
@juliusbrusseeDrift detector. Diff SPEC.md against code. Read-only, zero writes.
@juliusbrusseeSpare-budget design pass. Make one shallow module deep — smaller interface, behavior held, tests green before & after.
@juliusbrusseeIs this you? Claim your profilePut a plugin at the top of Plugins, from $9
