メニュー
Use when optimizing, benchmarking, packaging, comparing baselines, handling online score feedback, or coordinating parallel agents for a Huawei Algorithm Challenge 37 NSLB workspace identified by its solver and score-loop artifacts.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | ccdawn-huawei-nslb-score-loop |
| description | Use when optimizing, benchmarking, packaging, comparing baselines, handling online score feedback, or coordinating parallel agents for a Huawei Algorithm Challenge 37 NSLB workspace identified by its solver and score-loop artifacts. |
| license | MIT |
This is the Huawei NSLB project adapter for ccdawn-score-loop.
Use ccdawn-score-loop as the generic operating model, then apply the project-specific profile in references/huawei-nslb-profile.md when the request involves status, epoch work, workers, gate decisions, packaging, online feedback, failed-diff recovery, or project memory sync.
Project root: resolve from explicit `--project`, `CCDawn_HUAWEI_NSLB_ROOT`, or the current repository markers
Primary source: src/Solution.cpp
Tool: <codex-home>\skills\ccdawn-huawei-nslb-score-loop\scripts\score_loop_tools.py
Mutation space: references/mutation_space.json
Main ledger: docs/optimization-ledger.json
Online/offline weights: docs/online-offline-weights.json
Search graph: docs/search-graph.json
Attempt cards: docs/attempt-cards/
Failed diffs: docs/failed-diffs-index.json and docs/failed_diffs/
Known online-best anchor unless the project ledger says otherwise:
baseline: baseline-044-large-sparse-phase6-residual-target
online score: 344736301
source SHA256: 28b33d6d67d547fbc4415ec363a1c80db7a1f8e2fe1f308fd96a8a528a0441cb
child_result.json, attempt card, ledger/search-graph update, compiled diff, gate decision, or registered short submission package.ccdawn-score-loop, project-specific command, launch/recover workers, parent promotion gate, online feedback wait, project memory sync, or BLOCKED with one required input.ccdawn-brt 或更具体 owner，复合任务不吞其他 owner。下一步建议: <一个具体动作>。Before meaningful optimization, run or inspect the equivalent current outputs:
$project = if ($env:CCDawn_HUAWEI_NSLB_ROOT) {
(Resolve-Path -LiteralPath $env:CCDawn_HUAWEI_NSLB_ROOT).Path
} else {
(git rev-parse --show-toplevel 2>$null)
}
if (-not $project -or -not (Test-Path -LiteralPath (Join-Path $project "src\Solution.cpp"))) {
throw "请通过 --project、CCDawn_HUAWEI_NSLB_ROOT 或当前仓库提供 Huawei NSLB 项目根目录。"
}
$codexHome = if ($env:CODEX_HOME) { $env:CODEX_HOME } else { Join-Path $HOME ".codex" }
$tool = Join-Path $codexHome "skills\ccdawn-huawei-nslb-score-loop\scripts\score_loop_tools.py"
python $tool status --project $project
python $tool doctor --project $project --min-score 0
python $tool drift --project $project
Do not rely on chat memory as the source of truth for score, source hash, active workers, or package mapping.
Route these requests through the profile reference:
Use short upload names such as sub053.zip; store long metadata in docs/submission-map.json.
Workers must follow the generic ccdawn-score-loop contract plus these Huawei constraints:
src/Solution.cpp;child_result.json before completion.For the detailed worker prompt, command router, mutation families, risk-resurrection rules, online calibration rules, and memory sync commands, load references/huawei-nslb-profile.md.
After meaningful Huawei score-loop work, update project memory with the lane competition-huawei-nslb unless the user explicitly says to skip it.
Use when a user message needs Chinese-first intent inference, low-confidence collaborative clarification, routing, workflow weight control, skill choice, review/testing/planning/debugging/evaluation routing, proactive multi-thread collaboration or conflict coordination, continuation handling, execution permission inference, or protection from process over-escalation.
Use when active development or explicit refactoring has concrete code-structure risk: a source file is accumulating independently changing responsibilities, a class/component/function is hard to navigate or test, repeated changes touch distant sections, or the user asks to split a giant file; do not use for line count alone, generated/vendor/migration/schema/fixture files, or cohesive modules that remain easier to maintain together.
Use when the user explicitly enables an automated same-project multi-thread development loop that keeps peer Codex tasks progressing, recovers conflicts or stale coordination, integrates verified work into local main, and closes cleanup without repeated gates; do not use before opt-in, for one-off coordination, remote publication without permission, or subagent creation.
Use when existing independent Codex App threads work in the same project and need peer advice, collaboration proposals, shared progress awareness, ownership arbitration, conflict pause/resume, discussion, merge coordination, status exchange, or handoff through native thread tools without creating subagents.
Use when a verified or integrated software change has concrete temporary artifacts, generated noise, stale claims, merged local branches, or disposable worktrees to clean, or when the user explicitly asks to remove development residue or old branches.
Use when the user or project rules explicitly request project-local memory/dashboard initialization or updates, or when an existing `.docs/project-memory/` is needed for cross-session recovery, durable decisions, blockers, parallel-agent coordination, or formal handoff; do not auto-initialize or sync it for ordinary development.