user-shakedownlisted

# User shakedown Autonomous **ledger closure**. Where `/user-walkthrough` yields to the operator each turn, this skill drives itself: it works down the ledger's open items — verifying, quick-fixing, or logging each — until the ledger has zero open items. It is **pipeline-autonomous**: no confirmation prompts, and at wrap it EXECUTES the closers rather than offering them. It STOPS (by logging, never guessing) on anything that genuinely needs operator judgment. The ledger it reads and writes is defined once in [`.claude/references/shakedown-engine.md`](../../references/shakedown-engine.md) — this skill never redefines the row grammar, slug rule, or status vocabulary; it cites the engine. Because it uses the same slug rule, it resolves the **same** ledger file a `/user-walkthrough` may have started, and picks up mid-stream (load-before-derive per the engine — never re-seed). ## When to use / not - **Use:** UAT needs to be *closed out* on a just-built feature and you want the agent to do it autonomously — ideally armed under a `/goal` so it re-drives itself every turn until the ledger is clean. Good for draining a walkthrough's leftover open rows, or a fresh closure pass. - **Don't use** when you want to drive the exploration yourself (that is `/user-walkthrough`), when you have an already-clear script to just execute (`/user-uat`), or to refine a fuzzy UAT (`/review-uat`). ## Invocation ```text /user-shakedown <tool/feature> # autonomous closure of the named bu