메뉴
Adversarial review of prose drafts by N reviewers, then revise. Not for code.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Run the canonical Megaplan planning pipeline from its Arnold plugin home.
Launch one durable, mutation-authorized meta-fixer to repair a failed automated fixer and its missed backstop, retrigger ordinary repair, and prove the identified epic or session advances. Use when a watchdog, repair loop, meta-repair loop, or progress auditor ran or should have run but the real chain stayed stuck, repeated stale evidence, claimed false success, or failed to self-correct. Invoke with `--target "EPIC_OR_SESSION_TEXT"`.
Launch an external model as a subagent for a second opinion, adversarial review, or delegated work. Default pathway is an agentic DeepSeek / Kimi / Zhipu GLM hermes subagent (file/web/terminal tools, one process or fanned out N-wide); also Codex (GPT-5.5) and Claude via the Agent tool. Use for independent root-cause analysis, cross-checking your reasoning, judge/jury panels, or handing implementation to a different model.
Scaffold a new Arnold native-first pipeline package from the _template skeleton.
Survey every place loose work hides in a git repo — starting with the current checkout's own uncommitted/untracked/unpushed work, then local branches, all worktrees (`.megaplan-worktrees`, `.megaplan/bakeoffs/*`, agent-tool worktrees), stashes, detached HEADs, interrupted rebases/merges, submodules, remote branches on origin and other remotes, fork PR refs, sibling repo variants, other clones of this repo elsewhere on disk, the megaplan-cloud machine (Hetzner box) and every workspace on it, and GitHub Codespaces — classify each as land-on-main / delete / parked with reasoning, and act only on explicit per-item approval. Use when the user says "clean up loose branches", "prune branches", "what branches can I delete", "clean up worktrees", "review my stashes", "what's lying around in this repo", or asks for branch / worktree / stash housekeeping.
Steps to create a new Arnold native-first pipeline package from the template.
| name | writing-panel-strict |
| description | Adversarial review of prose drafts by N reviewers, then revise. Not for code. |
A multi-reviewer pipeline for rigorous prose revision. Three independent reviewers (pessimist, optimist, structuralist) critique a draft in parallel, a synthesis editor reconciles their feedback into a revision brief, and a revision editor produces the updated draft. A human gate lets you review the result and loop back for another round or stop.
writing-panel-strict is a native-default converted pipeline. Fresh runs
through megaplan run writing-panel-strict ... or
arnold pipelines run writing-panel-strict ... persist runtime ownership in
state.json.runtime_envelope.runtime and state.json.meta.executor. During
the M7 deprecation window, the derived graph remains available as a
compatibility fallback: pass --runtime graph (or the deprecated
--executor graph) for a fresh run that must use the graph executor. Existing
graph-born plan directories keep resuming on graph. Native-born runs resume on
native, and corrupt native cursors fail closed rather than silently falling
back to graph.
| Mode | Description |
|---|---|
polish | Light editing — tighten prose, fix errors, preserve structure |
restructure | Reorganize sections, improve flow, may move content |
provoke | Push the draft to be bolder, more opinionated, less safe |
draft (file, required): Path to the markdown draft to review.# Run with default profile
megaplan run writing-panel-strict path/to/draft.md
# Run with a specific mode
megaplan run writing-panel-strict path/to/draft.md --mode polish
# Run with a specific profile
megaplan run writing-panel-strict path/to/draft.md --profile @writing-panel-strict:standard
# Resume after human gate
megaplan resume <plan-dir> --choice continue
megaplan resume <plan-dir> --choice stop
continue to loop back for another round, or stop to finish.| Profile | Description |
|---|---|
@writing-panel-strict:standard | Default — Claude low-effort for all stages |
@writing-panel-strict:premium | Coming soon — Claude high-effort reviewers |
@writing-panel-strict:cheap | Coming soon — DeepSeek for all stages |