JuliusBrussee

81 mods in 3 plugins, 7 repositories, 116k stars between them.

caveman

01

JuliusBrussee/caveman

Plugin Claude Code

Bundles 22 skills, 9 commands, 3 agents · 1,297 tokens together

Ultra-compressed communication mode. Shortens prose while preserving technical terms, code, commands, and exact errors.

not rated 113k +1.8k changed 2d ago A tokens not measured original Apache-2.0

caveman

02

JuliusBrussee/caveman

Plugin Gemini CLI

Ultra-compressed communication mode. Shortens replies while preserving technical substance, code, commands, and exact errors.

not rated 113k +1.8k changed 2d ago A tokens not measured original Apache-2.0

ck

03

JuliusBrussee/cavekit

Plugin Claude Code

Bundles 7 skills, 7 commands · 915 tokens together

Cavekit — compressed spec-driven dev. Full loop (grill → spec → research → review → build) over one SPEC.md file: three core commands + four reach-for. Caveman encoding. Bug-to-spec backprop.

not rated 1.2k +1 2mo ago A tokens not measured original MIT

critic

04

JuliusBrussee/caveman-code

Agent Claude Code

Adversarial review of a plan or implementation. Pokes holes; assumes the author was over-confident.

not rated 941 +1 2mo ago A 24 tokens original MIT

explore

05

JuliusBrussee/caveman-code

Agent Claude Code

Read-only codebase reconnaissance. Returns a compressed inventory another agent can act on without re-reading the tree.

not rated 941 +1 2mo ago A 24 tokens original MIT

implementer

06

JuliusBrussee/caveman-code

Agent Claude Code

Execute a concrete plan or patch description by editing files in an isolated git worktree.

not rated 941 +1 2mo ago A 20 tokens original MIT

reviewer

07

JuliusBrussee/caveman-code

Agent Claude Code

Critique a diff or a slice of code. Returns prioritized findings with file:line citations and suggested fixes.

not rated 941 +1 2mo ago A 25 tokens original MIT

tester

08

JuliusBrussee/caveman-code

Agent Claude Code

Run the project's test suite, parse failures, and report root cause + suggested fix per failing test.

not rated 941 +1 2mo ago A 22 tokens original MIT

clean

09

JuliusBrussee/caveman-code

Command Claude Code

Remove obvious dead code, unused imports, and stale TODOs.

not rated 941 +1 2mo ago A 15 tokens original MIT

commit

10

JuliusBrussee/caveman-code

Command Claude Code

Stage and commit the current working tree with a conventional-commit message.

not rated 941 +1 2mo ago A 16 tokens original MIT

explain

11

JuliusBrussee/caveman-code

Command Claude Code

Explain a file, function, or concept in the codebase.

not rated 941 +1 2mo ago A 16 tokens original MIT

fix-types

12

JuliusBrussee/caveman-code

Command Claude Code

Run the project's type-checker and fix the smallest set of errors.

not rated 941 +1 2mo ago A 17 tokens original MIT

goal

13

JuliusBrussee/caveman-code

Command Claude Code

Start an autonomous goal loop in the background. Usage. /goal.

not rated 941 +1 2mo ago A 19 tokens original MIT

log

14

JuliusBrussee/caveman-code

Command Claude Code

Summarise recent git history.

not rated 941 +1 2mo ago A 9 tokens original MIT

migrate

15

JuliusBrussee/caveman-code

Command Claude Code

Walk through an opinionated multi-step code migration.

not rated 941 +1 2mo ago A 13 tokens original MIT

perf

16

JuliusBrussee/caveman-code

Command Claude Code

Profile or analyse a hot path and propose a measured speedup.

not rated 941 +1 2mo ago A 15 tokens original MIT

review

17

JuliusBrussee/caveman-code

Command Claude Code

Review the diff on the current branch against main (or a target branch).

not rated 941 +1 2mo ago A 17 tokens original MIT

sec-review

18

JuliusBrussee/caveman-code

Command Claude Code

Security review of the current diff (or the whole repo, if --all).

not rated 941 +1 2mo ago A 19 tokens original MIT

test

19

JuliusBrussee/caveman-code

Command Claude Code

Run the project's test suite and triage failures.

not rated 941 +1 2mo ago A 12 tokens original MIT

JuliusBrussee/caveman-code

Skill Claude Code

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, or auditing UI code against design tokens.

not rated 941 +1 2mo ago A SkillSpector: pass 62 tokens original MIT

cavekit-methodology

22

JuliusBrussee/caveman-code

Skill Claude Code

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.

not rated 941 +1 2mo ago A SkillSpector: pass 60 tokens original MIT

cavekit-revision

23

JuliusBrussee/caveman-code

Skill Claude Code

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 convergence stalls, or when the same class of bug keeps…

not rated 941 +1 2mo ago A SkillSpector: pass 73 tokens original MIT

JuliusBrussee/caveman-code

Skill Claude Code

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, or debugging why an agent is producing output that nobody…

not rated 941 +1 2mo ago A SkillSpector: warn 64 tokens original MIT

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: