ArchestackBest of Vibe Coding: List on GitHub ↗ (opens in a new tab)

#402402nd of 463 in Skills, plugins and prompts

cavekit

Spec-driven build loop for Claude Code using one compressed SPEC.md file

Repository on GitHub ↗ (opens in a new tab)
Stars
1.2k
License
MIT
Last commit
Aug 2026
Last release
Apr 2026

Missed the latest GitHub check; these facts may be older.

Overview

Cavekit installs nine skills and seven /ck: slash commands into Claude Code. They run a loop of grill, spec, research, review and build over a single SPEC.md at the repo root, written in a token-compressed format, with no sub-agents. Test failures are written back into the spec as bug entries and invariants. The project is marked frozen as of August 2026, with development moved to caveman.

Who it is for: Claude Code users wanting a lightweight spec-driven workflow

Strengths

  • Single SPEC.md survives context resets; plain markdown with pipe tables
  • Nine skill descriptions cost about 1.1k context tokens, per README
  • Failed tests are written back to the spec as §B bugs and §V invariants
  • Right-sizing: a one-line fix needs only /build, not the full chain

Weaknesses

  • Frozen since August 2026: no new features or fixes expected
  • Claude Code only; no stated support for other coding agents
  • No sub-agents or parallel workers; the older v3.1.0 line is needed for those
  • Token savings figures (75%, 16x lighter than spec-kit) are the author's own claims

What it needs

  • no GPU
  • Needs Claude Code, Node.js (npx skills CLI)
  • Models: Claude

Also in Skills, plugins and prompts

See all 463
Also in Skills, plugins and prompts
RankProjectScore
1claude-memPersistent cross-session memory plugin for Claude Code and other coding agents116.4k stars+16,587 (+17%) in 30 daysApache-2.097 out of 100
2impeccableDesign skill with 24 slash commands for AI coding agents' frontend output Live demo ↗ (opens in a new tab)108.8k stars+28,464 (+35%) in 30 daysApache-2.097 out of 100
3gstackMarkdown slash-command skills that give Claude Code planning, review, QA and ship roles141.3k stars+5,544 (+4.1%) in 30 daysMIT96 out of 100
#4rufloAgent harness adding swarms, memory and hooks to Claude Code and Codex Live demo ↗ (opens in a new tab)81.5k stars+7,374 (+9.9%) in 30 daysMIT96 out of 100
#5cavemanShortens coding-agent replies and compresses tool output before the model reads it Live demo ↗ (opens in a new tab)119.3k stars+8,422 (+7.6%) in 30 daysApache-2.095 out of 100