Use when refilling the Fleet feature queue — decompose one signed-off goal or one section of an existing PLAN into atomic, independently-mergeable fleet:pending issues. Recon before decompose; never write implementation code.
Scanned 9/2/2026
Install to Claude Code
npx -y skills add fagemx/edda --skill fleet-epic-split --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Fleet Epic Split?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/fagemx-fleet-epic-split)More formats (shields.io, HTML) on the badges page.
---
name: fleet-epic-split
description: Use when refilling the Fleet feature queue — decompose one signed-off goal or one section of an existing PLAN into atomic, independently-mergeable fleet:pending issues. Recon before decompose; never write implementation code.
---
# Fleet Epic Split(出題器 A)
把一顆已圈的目標(或一份既有 PLAN.md 的一節)拆成互相獨立、各自可 merge 的原子任務單。
你**只起草提案**(fleet:pending),撕成 ready 是操作者的事。慣例見 repo 自身的 `CLAUDE.md`/`AGENTS.md`(本 repo:`.claude/CLAUDE.md`)。
## 開工前
- repo 根有 `FLEET_PAUSE` → idle 退出。
- 確認目標:一顆 `fleet:goal` issue,或操作者指定的 PLAN.md 一節/一個 epic 描述。
## 三段式(順序不可跳)
### ① recon(只准調查)
讀 repo 現況與目標,產出帶 **file:line 證據**的現況報告:目標要的東西,哪些**已經做完**(查
既有程式、測試、PLAN 的 `[x]` 標記與實際碼是否相符——標了不算數,要碼在),哪些還沒。
**禁止**在這一段提解法或寫計畫。已完成的部分不生單。
### ② decompose(拆原子單)
把「目標 − 已完成」的差距拆成原子單。每張**必過獨立性三問**:
1. 單獨 merge 會不會弄壞 main?
2. 跟別張單改同一批檔案嗎?
3. 不看其他單,能不能直接開工?
兩張搶同一檔 → 合併成一張,或標 `獨立性: blocked by #N` 明示依賴。
**禁止**:出「重構整個 X」這種巨單;出一個 session 做不完的單(做不完就再拆)。
### ③ propose(照模板起草)
每張單照六欄寫成 issue body(背景/改哪裡/doneWhen/verify/獨立性/尺寸),缺欄不准發。
建單:`gh issue create --title "<動詞開頭>" --body-file <tmp> --label fleet:pending,lane:feature`。
**每輪上限 30 張**。全部產完,回報清單(issue 號+標題+獨立性)給操作者裁決。
## 界線
你不撕 ready、不寫實作碼、不 merge。你的產出是「待簽的提案」,下游有測試閘門兜底,
但錯的**方向**沒有閘門攔得住——所以寧可少拆、標清依賴,也不硬湊數量。
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!
Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...
Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.
**Complete production-ready guide for Google Gemini embeddings API** This skill provides comprehensive coverage of the `gemini-embedding-001` model for generating text embeddings, including SDK usage, REST API patterns, batch processing, RAG integration with Cloudflare Vectorize, and advanced use cases like semantic search and document clustering. ---
Interview, source-challenge, verify, save, and ADR-gate fuzzy coding requests into Codex-ready implementation specs. Use when a feature, bugfix, refactor, migration, repo-wide change, or architecture task needs user-verified requirements, source-backed decisions, durable architecture decisions, acceptance criteria, validation commands, rollout notes, saved spec/ADR files, and a Codex execution prompt. Do not use when already fully specified or when the user wants direct implementation now.
Use when a repo needs CodeGraph plus ast-grep for Codex MCP setup, exploration, impact analysis, structural search, or safe refactor planning.