juliusbrussee/cavekit

4812

caveman Caveman encoding for SPEC.md and spec-adjacent writes. Loaded by /spec, /build, /check. Cuts tokens ~75% vs prose whi…

juliusbrussee/cavekit

2.7K

4947

build Plan-then-execute implementation against SPEC.md. Native single-thread loop, no sub-agents. On test or build failure,…

juliusbrussee/cavekit

2.7K

4963

spec Create, amend, or backprop bugs into SPEC.md at repo root. Sole mutator of the project spec. Triggers when the user a…

juliusbrussee/cavekit

2.6K

4976

check Read-only drift detector. Diffs SPEC.md against current code and reports violations grouped by severity. Writes nothi…

juliusbrussee/cavekit

2.6K

5114

backprop Bug → spec protocol. When a bug is found or a test fails, trace the cause, decide whether a new §V invariant would ca…

juliusbrussee/cavekit

2.6K

6

grill Calibrated interrogation of a fuzzy idea before it becomes a spec. Asks one question at a time, recommends an answer,…

juliusbrussee/cavekit

727

7

research Gather external knowledge the spec needs and distill it into §R — the durable research log — so build grounds in fact…

juliusbrussee/cavekit

725

8

review Adversarial senior review of the spec before any code is written. Constructs a skeptical reviewer whose authority com…

juliusbrussee/cavekit

723

9

deepen Optional design-improvement pass for when you have spare usage to drain. Finds the shallowest modules in the code the…

juliusbrussee/cavekit

710

10

design-system How to write and maintain a DESIGN.md in the 9-section Google Stitch format. Covers the 9-section structure, design t…

juliusbrussee/cavekit

18

11

context-architecture Progressive disclosure architecture for organizing project context as a DAG (directed acyclic graph). Agents enter at…

juliusbrussee/cavekit

17

12

convergence-monitoring Detecting whether agent iterations are converging toward a stable solution or hitting a ceiling. Covers convergence s…

juliusbrussee/cavekit

17

13

documentation-inversion Inverts the traditional documentation flow from code-to-wiki-for-humans (which rots) into code-to-CLAUDE.md-to-skills…

juliusbrussee/cavekit

17

14

methodology Core Cavekit methodology — the master skill that teaches the Hunt lifecycle and routes to all sub-skills. Covers the …

juliusbrussee/cavekit

17

15

ui-craft Authoritative guide for implementing stunning, accessible, performant UI. Synthesizes design engineering philosophy, …

juliusbrussee/cavekit

17

16

brownfield-adoption Step-by-step process for adopting Cavekit on an existing codebase. Covers the 6-step brownfield process, bootstrap pr…

juliusbrussee/cavekit

16

17

cavekit-writing How to write Cavekit-quality kits that AI agents can consume effectively. Covers implementation-agnostic cavekit desi…

juliusbrussee/cavekit

16

18

peer-review Patterns for using a second AI agent or model to challenge the primary builder agent's work. Covers six review modes …

juliusbrussee/cavekit

16

19

revision Trace bugs and manual fixes back to kits and prompts, then fix at the source so the iteration loop can reproduce the …

juliusbrussee/cavekit

16

20

speculative-pipeline A pipeline execution strategy where downstream stages start before upstream stages finish, using staggered timing wit…

juliusbrussee/cavekit

16

21

impl-tracking Implementation tracking documents for maintaining living records of what was built, what is pending, what failed, and…

juliusbrussee/cavekit

15

22

prompt-pipeline How to design the numbered prompt pipeline that drives Hunt phases in Cavekit. Covers greenfield 3-prompt patterns, r…

juliusbrussee/cavekit

15

23

validation-first Validation-first design for AI agent output — every spec requirement must be automatically verifiable. Covers the 6-g…

juliusbrussee/cavekit

15

24

peer-review-loop Peer Review Ralph Loop — combines Cavekit kits with a Ralph Loop and true cross-model peer review using Codex (OpenAI…

juliusbrussee/cavekit

11

25

caveman-internal Internal token-compression protocol for Cavekit agent artifacts — handoff notes, artifact summaries, context bundles,…

juliusbrussee/cavekit

5

26

complexity-detection Classify a task or feature as quick, standard, or thorough so the rest of the pipeline (budget sizing, model routing,…

juliusbrussee/cavekit

5

27

autonomous-loop How Cavekit's autonomous execution loop works — state machine, stop hook, completion sentinel, lock, budgets, iterati…

juliusbrussee/cavekit

4

28

capability-discovery Detect which MCP servers, Claude Code plugins, and CLI tools are available in the current environment, so kits and bu…

juliusbrussee/cavekit

4

29

graphify-integration Optional knowledge-graph integration. When `graphify-out/graph.json` is present, architect/researcher/reviewer/task-b…

juliusbrussee/cavekit

4

30

karpathy-guardrails Behavioral guardrails for Cavekit agents. Four principles — think before coding, simplicity first, surgical changes, …

juliusbrussee/cavekit

4