← ClaudeAtlas

boardsweeplisted

True-up every agent-authored live-board component — verify done entries against the repo's actual history, migrate finished planned items, close resolved decisions, correct current, prune the registry, refresh the wiki projection, normalize tone/schema. Excluded by contract: the realtime chat, user-authored slots, and mode/policy texts. Retrospective accuracy only: this skill fixes what the board claims; /roundnext decides what happens next. Invoke on a cadence, after a batch of work landed off-board, or whenever the board smells stale.
SoliEstre/EstreGenesis · ★ 8 · AI & Automation · score 78
Install: claude install-skill SoliEstre/EstreGenesis
# /boardsweep — one full board true-up (every agent-authored component) The complement of `/roundnext` on the time axis: roundnext is **forward-looking** (which planned item should start, which blocker to attack), boardsweep is **retrospective** (is what the board currently says actually true?). A board that isn't swept accumulates ghost claims — done entries whose identifiers resolve to nothing, planned items that quietly shipped, decisions that stay "open" after the work answered them, a current lane describing last week. A sweep changes recorded facts; it never starts work. ## 0. Load board truth + ground truth - **Board** ��� read the live state (`board_state_get` MCP tool, or the board's `state.json` directly). The sweep set is **every agent-authored component**: `done[]`, `planned[]`, `current[]`, `decisions[]`, the project registry, the wiki/vocabulary projection where the board carries one, and the meta fields (`updatedAt` etc.). - **Excluded — never rewritten by a sweep**: - the realtime chat / per-channel message history (in the reference runtime, the `ws-history` store and everything the conversation channels carry) — that layer is a *record of what was said*, not a claim to true-up; - user-authored slots (free-request fields, feedback queues) — user speech is not the agent's to edit; - the board's mode/policy texts — the sweep **obeys** whatever policies they carry (content tone, grouping, routing or the workspace's equivalents) and may flag a violation,