wrap-up
Skill547 starsChanged 9 days agoMITView on GitHub
只在使用者明確要求 session 收尾、整理專案記憶、或 commit memory 時使用。不因一般任務完成、狀態報告或程式碼 commit 自動觸發。
- Commits and pushes
64 other projects use a version of this.
Adopted by
Most recently taken up by
- flegaspi700/pmo-project-standardizationtheir own version, first seen Nov 16, 2026
- manavpurswani/claude-wrap-uptheir own version, first seen Nov 16, 2026
- mm-arturo/claude-code-skillstheir own version, first seen Nov 15, 2026
- PaulRBerg/agent-skillstheir own version, first seen Nov 13, 2026
- bas1l/claude-skillstheir own version, first seen Nov 11, 2026
- adeonir/agent-skillstheir own version, first seen Nov 10, 2026
Used with
Projects with it also use new-project, refine and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
zcagSkillnew-project
Guide users through creating a new software project from scratch — from raw idea to actionable tasks. Use this skill whenever someone says "new project", "start a project", "create a project", "scaffold a project", "I want to build X", "help me plan a project", "I have an idea for an app/tool/service", or mentions they're starting something new. Also trigger when someone asks for help with project planning, tech stack decisions, MVP scoping, feature planning, or figuring out what to build and how to break it down. When the user describes any project idea and wants to move forward on it, this skill should activate.
Also in 3 projects that use this
AlexPEClubSkillrefine
Always use when the user wants to discuss an exsiting feature or specification. Open an existing feature spec to improve, extend, or fundamentally challenge it. Pass the feature ID as argument (e.g. /refine PROJ-2).
Also in 3 projects that use this
MicrosoftSkilldoctor
Diagnose an Amplifier CLI environment through the CLI doctor agent. USE WHEN a user requests CLI runtime, settings or missing-capability inspection. DO NOT USE for project bugs or other hosts.
Also in 4 projects that use this
austintgriffithSkillorchestration
Use when a Scaffold-ETH 2 build leaves localhost: planning the launch order, deploying and verifying contracts on a live network, standing up a fork of a real chain to rehearse a deploy, or fixing a contract bug that is already onchain. Not for running the app or a demo against that fork (`/frontend-playbook`).
Also in 3 projects that use this
MicrosoftSkillsync
Sync the current session branch with its upstream branch, or publish the current session branch to a remote. Use when the user asks to sync a branch, pull latest changes, rebase onto upstream, push current branch, publish branch, or set upstream.
Also in 3 projects that use this
nicolasestremSkillbrainstorm
Structured idea generation and solution exploration with systematic analysis.
Also in 6 projects that use this
- +1 more from microsoft
Other versions64 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 64
- rohitg00/pro-workflow★ 2.9ksame name
- anthropics/launch-your-agent★ 1ksame name
- codewithmukesh/dotnet-claude-kit★ 808same name
- aspi6246/Claude-Code-Skills-for-Academics★ 153same name
- PaulRBerg/agent-skills★ 104same name
- KerberosClaw/kc_ai_skills★ 79same name
- jonathanmalkin/jules★ 65same name
- simoncorry/foundry★ 39same name
- alxyrgin/health-os★ 38same name
- mshadmanrahman/claude-for-designers★ 32same name
54 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository12 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- public/llms.txtllms.txt
- supabase-optimizeSkill
- wrap-upSkill
- accessibility-analysisSkill
- blog-writingSkill
- layer-onboardingSkill
- service-coverageSkill
- supabase-optimizeSkill
- three-3d-componentSkill
- weekly-auditSkill
The file
We split this skill into the rules it gives an agent
1 rule. 0 are common, 1 is distinctive, and 0 are very specific to this project.
0
Common
Most of the 60 edited versions we read keep these.
1
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 81
Commit 前先記錄並列出 git diff --cached --name-only 的 cached path set,辨識並保留 unrelated pre-staged paths;不要求整個 staging area clean。
“Common” is measured: kept by at least 6 in 10 of the 60 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Wrap-up v2
把本 session 收成可審查、可接手、不誤報 release 狀態的 checkpoint。 Wrap-up 只處理記憶與收尾 commit;不因此取得 upload、deploy、push、merge 或資料變更授權。
1. Gather:先定義 scope
先用對話與工具記錄建立 scope ledger:
- touched repos,每個 repo 的 current branch、upstream、intended base、
git status。 - 本 session 的 commit range;比較 intended base,不假設 base 名為
master。 - external systems 與真實 side effects:object storage、DB/migration、deploy platform、CDN、PR。
- 非本 session 的 dirty files 或 commits;標示後保留,不代為整理。
然後讀 .claude/memory/README.md 的 routing。依本 session 事件選讀可能要更新的 memory;
選中的檔案要讀完,但不固定讀 9 檔,也不整包讀 .claude/memory/。
證據只來自本 session 對話、repo 現況、實際命令與已維護文件。不跨 session 臆測,不把 commit message 當 runtime 證明。
2. Release truth matrix
只要本 session 碰到 artifact 或 release,固定用下列順序記錄每個 release unit:
| build | contract/wire | stage | upload | readback | pull | deploy | HTTP | browser |
每格只能是 done / failed / blocked / unknown / not run / N/A,並附最小可核對證據。
unknown 只用於證據不足、無法判定真實狀態;已知卡點用 blocked,尚未執行用 not run。
- 上游完成不等於下游完成;
upload ≠ deploy ≠ browser verified。 - upload 後要分開記 readback(checksum / size / metadata)。
- 自動測試記指令、範圍與結果;不把 tsc/unit tests 寫成視覺驗收。
- browser 或 network 工具不可用時寫
blocked,不改用代碼推斷補勾。 - 非 release 任務可省略此矩陣。
Contradiction gate
若對話、git、feature docs、memory 或 external state 相互衝突:
- 先列出衝突與影響,用當下可得的第一手證據核對。
- 無法解決就保留
blocked/unknown,不為收尾而選好看的答案。 - 在矛盾澄清前,不將相關項目寫成 done,不進入 Apply。
3. Analyze 與 routing
依 memory/README.md 的當前定義分流,不為了「全更」而碰無關檔案。
docs/PROJECTS.md:更新相關專案列與「最後盤點」日期(發版時docsFreshness測試會檢查)。STATUS.md:rewrite,只保留當前 touched repos、release truth、blockers 與下一步。DATA_SCOPE.md:數字必須直接從 canonical data、DB query、artifact metadata、wc/parser 或同等第一手來源驗證;不單信對話摘要或舊文件。BACKLOG.md:每個未完成項都要有 blocker、next step 與 acceptance criteria;不以「待處理」結案。INCIDENTS.md/REFLECTIONS.md:append-only;沒有新證據不回寫舊條目。PRINCIPLES.md:新舊原則衝突時,明確取代舊原則並留演進原因。PLAYBOOKS.md:只記已重複實證的流程。單次做法放 feature docs 或 reflection。
收尾必須留下 next-session entry:目標 repo/branch、當前 blocker、第一個可執行步驟、驗收條件。
4. Draft 與 Confirm
先只回報:
- scope ledger 摘要。
- release truth matrix(若適用)。
- memory 變更總表:每檔一行、一句摘要(≤20 字)。
- 衝突、未完成項、next-session entry 與不應碰的檔案。
不 dump 完整 diff、markdown 全文或每個 edit 片段。問:全採用 / 看某檔細節 / skip 某檔? 使用者確認前不編輯 memory,不 stage,不 commit。
5. Apply 與 atomic commits
只在 Confirm 後套用核准項目:
- 編輯前重讀目標檔;保留 unrelated dirty changes。
- 若同一 target memory file 已含非本 session 或平行 session hunks,path-scoped commit 無法隔離;立即停止並請使用者協調,不得代為 commit 整檔。
- Commit 前先記錄並列出
git diff --cached --name-only的 cached path set,辨識並保留 unrelated pre-staged paths;不要求整個 staging area clean。 - 每個 memory 檔做 path-scoped diff-check 與
git add <exact-path>,再用git commit --only -m "..." -- <exact-path>建立一檔一個 atomic commit。禁用git add -A。 - commit 訊息用
memory: <action> <file> (<summary>),不加假 co-author 或模型身分。 - pre-commit hook 失敗時,commit 尚未產生:修正後重跑同一個 commit,不新開 fix commit,不 amend 不存在的 commit。
STATUS.md最後提交。- 結束時只確認 target memory paths 已 clean,並列出仍保留的 unrelated staged 與 dirty state。
Push 邊界
不自動 push。最後顯示 current branch、upstream 與 ahead/behind,單獨詢問 push 授權。
只提示 current branch 的明確 refspec(例如 git push origin HEAD:<current-branch>),
不寫死 master/main,不把 push 授權延伸成 PR、merge 或 deploy 授權。
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

