From ck
Plan-then-execute implementation against SPEC.md, with auto-invoked backprop on failure. Runs native single-thread loop, no sub-agents. Useful for structured, specification-driven development.
How this skill is triggered — by the user, by Claude, or both
Slash command
/ck:buildThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Single-thread native plan→execute. You are main Claude. No swarm.
Single-thread native plan→execute. You are main Claude. No swarm.
SPEC.md. If missing → tell user to invoke the spec skill first. Stop.FORMAT.md once if not loaded.§T.n → that task only--next → lowest-numbered row with status . or ~--all or empty → every . row in §T orderHigh blast radius (shared module, auth, data, money, public §I)? Run /review first. Trivial & reversible? Skip planning ceremony, just do step EXECUTE.
Native plan mode — you delegate to it, you do not reinvent task breakdown. For chosen task(s):
Show plan. Wait for user OK unless auto mode.
Per task in order:
. → ~. Just write to SPEC.md.~ → x. Next task.On test/build failure:
bug: <cause> first, let it update §V and §B, then resume build against updated spec.Rule: never silently fix root-cause without considering backprop. §B is the memory that stops recurrence.
T<n>: <goal line> + §V cites.Task x only if:
cat SPEC.md | grep §T is the dashboard.npx claudepluginhub juliusbrussee/cavekit --plugin ckExecutes approved implementation tasks from a plan.json file created by spec-plan. Tracks progress via beads or harness todos, enforces TDD, and reports between batches. Invoke when ready to start coding from an approved plan.
Converts approved specs into bite-sized TDD implementation plans with exact file paths, code snippets, and commit boundaries. Use after brainstorm-lean when implementation has not started.
Provides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.