本文へ移動
cccskills
無料GitHub で公開

os-big-picture

ALWAYS invoke this skill when the user asks where the project as a whole stands - "where are we", "what's the big picture", "what have we built", "what is in this project", "map the project", "what does this thing even do", "what is stale", "what can we delete", "update the roadmap" - in any language, and whenever a session report has just been written. It keeps one file, BIG-PICTURE.md: what the product is, every feature with how far it got, which parts nobody has touched, and what is queued next. Age and wiring are measured from git; how far a feature got comes from the reports or says "not checked". Where a tracker is connected it offers to open the queue as tickets, after a yes and never before. Never invents work, never deletes code.

インストール方法を見る

含まれるファイル(4)

  • SKILL.md7.4 KB
  • references/01-quiet-is-not-dead.md3.7 KB
  • references/rationale.md6.5 KB
  • scripts/census.sh3.9 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

os-big-picture

One file, BIG-PICTURE.md, where os-whats-next already looks for a backlog. It answers "where are we now" - not a plan, nothing in it a promise. Write it in the user's language; paths stay English.

When to use it. The user asks where the project stands, or there is no BIG-PICTURE.md yet → the full pass: describe, census, signals, backlog. A session report was just written → the fold-in: only the rows that session touched. Both re-measure, because a session that moves no row is the one after which the dates go quietly stale.

Step 1 - find the file, and never clobber it

BIG-PICTURE.md in the project root, and it may already be somebody's own work. Everything this skill writes lives between <!-- open-steps:begin --> and <!-- open-steps:end -->; no markers → append at the end; no file → create it, and add it to this clone's ignore list, never to .gitignore:

f=BIG-PICTURE.md; e="$(git rev-parse --git-dir)/info/exclude"
git check-ignore -q "$f" || git ls-files --error-unmatch "$f" >/dev/null 2>&1 \
  || grep -qxF "$f" "$e" 2>/dev/null || printf '%s\n' "$f" >> "$e"

Each of the three is a reason to leave it alone. Say in one line that you did it; git add -f BIG-PICTURE.md shares it instead. Never read Last worked on or Signal back out of the file - only Stage and the queue carry forward.

Step 2 - the census, measured by the script

bash ${CLAUDE_SKILL_DIR}/scripts/census.sh .

Where your tool does not set ${CLAUDE_SKILL_DIR}, run it from the project root as bash <this skill's folder>/scripts/census.sh .. If it cannot run at all, say so and mark every measured column "not checked"; never estimate them.

An AGE line, then one measured PART row per part: path, last worked on, commits in six months, whether anything outside reaches it, the signal.

Step 3 - reading what it printed

SignalWhat it meansWhat to do
activeBeing worked onNothing
stableQuiet, and something still reaches itLeave it alone - finished and in use
unused - N monthsQuiet, and nothing reaches itThe only real retire candidate

When the AGE line says young, write one line directly above the table, with the day count and date from the AGE line: "This project is N days old. Nothing here can be quiet for six months yet, so the Signal column starts to mean something on DATE." Every active is then a fact about the calendar, not the code. On an ok line, write no such note.

Step 4 - the backlog, sourced and never invented

Three sources, each item naming its own: session reports in ~/.claude/open-steps/reports/<project>/ (⏳ deferred rows, "Anything needed from you", recorded debt); Step 3 (each unused part, one retire candidate); the user, by hand. Nothing else - a gap you noticed while reading the code is not a task, so say it in the chat and let the user decide.

Step 5 - the tracker, if there is one

An offer, never a quiet action, and only for the "What is next" rows.

  1. Find it, never install it. gh repo view --json hasIssuesEnabled -q .hasIssuesEnabled, then any tracker already connected. Nothing connected is a normal answer: say so under "What is next", then Step 6.
  2. Search before you propose: gh issue list --search '<a few words>' --state all --json number,title,state. A closed ticket is still a match; where the search did not run, say so.
  3. Ask, then create. Plain lines through os-ask-simple; on a yes create those and nothing else, and only where this pass measured what the item rests on. Write each number back into its row, same pass.

Step 6 - plain words, before it is written

Put every sentence in the block through the os-say-simple rules first: one sentence one idea, active voice, one word for one thing. That covers the prose, the "What it does" column and the backlog items - never a measured cell.

The shape

<!-- open-steps:begin -->
_Measured <date>. Stages are dated where they stand; anything undated in this
file is not measured._

## What this is

<Three or four sentences. What the product does, who runs it, what it runs
on. No jargon - a reader who has never seen the code.>

## What is in it

| Feature | What it does | Stage | Last worked on | Signal |
|---|---|---|---|---|
| <plain name> | <one line> | live (1 Sep) | 1 Sep | active |
| <plain name> | <one line> | not checked | 28 May | stable |

## Worth retiring

<One line each: what it is, how long unused, what would break. Empty is a
result - write "nothing found" and keep the heading.>

## What is next

- <item> - <source>
- <item> - <source> - #<ticket number, once one exists>

_Age and wiring measured <date>, fresh this pass. Stage comes from session
reports and is only as current as the date beside it._
<!-- open-steps:end -->

Stage is one of building, built - not shipped, live, retired or not checked, and it carries the date it came from, always - live (12 Jul), never bare live, and not checked never dated. When the newest date in that column is months behind the newest commit, the reports have stopped: say so.

Hard rules - these rules are the skill

  1. Measured and assumed never mix. Last worked on and Signal are script output every time. Stage has no measurement.
  2. Enumerate, never guess. The parts come from the script.
  3. Quiet is not dead. No retire candidate on age alone: quiet plus wired in is stable, which is what stops the map recommending you delete a working product.
  4. Never invent work. Every backlog item names its source.
  5. Never clobber. Only between the markers. Their file stays theirs.
  6. Recommend, never act. No deleting code, moving files, opening a pull request. "Worth retiring" is a sentence, not an action.
  7. Plain words. A feature is named as the user says it, not as the folder.
  8. One screen per section. Twelve features is a wall - group and say so.
  9. A ticket only after an explicit yes. Rule 6 is the code, this is everyone else's inbox.
  10. The plain-words pass never touches a measured cell.
  11. Re-measure, never quote. A number read out of the file is a number about the past wearing today's date.
  12. Every claim carries its date or says it has none. A map that cannot go visibly stale will go invisibly stale.

Why the rules are there, the traps the script cannot catch, and a worked example: references/.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

ALWAYS invoke this skill before asking the user any technical question or offering options, and whenever they ask to be asked in plain words - "ask simple", "ask me simply", "ask me in plain words" - in any language. ALWAYS invoke it too when the user asks for your view on a technical choice: whether it is worth doing, more than the problem needs, or replaceable by something simpler, and which option you would take. Invoke it even when the code makes the answer look obvious: the checks are what make the answer more than a guess. Rewrites the question in plain words and always ends with one marked recommendation. A structural choice first passes six checks, shown as a table: effort now, simpler substitute, extra work later, lock-in, over-engineering, easy to undo. Doing nothing is always weighed.

日本語の概要は準備中です。原文の説明を表示しています。

kharmanskyi/open-steps1,2842026年10月6日 更新

ALWAYS invoke this skill when the user asks about work done outside this session - "check work", "check the others", "check other sessions" - or to accept one: "the session is done, check it", "can we merge it" - in any language. You are the receiving party: treat the report as a claim, verify each part against machine state, name every gap between claimed and true. Verified-ready work merges in the same pass. End with what is left, one next step, and any reply another session needs, ready to paste.

日本語の概要は準備中です。原文の説明を表示しています。

kharmanskyi/open-steps1,2842026年10月6日 更新

ALWAYS invoke this skill when work wraps up or the user asks how it went - "done or not", "are we done", "what happened", "report", "what's the status of this ticket" - in any language, and when a Stop hook asks for a session report. Produces a ten-line plain- language report: a lead, a checkmark table, and a verdict - fully done, anything needed from you, new debt, safe to close. Every "yes" names its proof; unverified says "not checked". Saves the report so the next session starts from it instead of re-exploring the repo.

日本語の概要は準備中です。原文の説明を表示しています。

kharmanskyi/open-steps1,2842026年10月6日 更新

ALWAYS invoke this skill when the user asks for simpler or shorter about something said or written - "say it simply", "what does this mean", "I don't understand your answer", "too long", "wait, what?", "bro" - in any language, about any text: your own answer, a report, a review comment, an error. "I don't understand what to DO" is os-step-by-step; this skill restates text. It restates for a reader who does not read code: leads with the point, keeps every number, warning and caveat - no facts added, no bad news dropped. A number returns exactly that many points, most important first.

日本語の概要は準備中です。原文の説明を表示しています。

kharmanskyi/open-steps1,2842026年10月6日 更新

ALWAYS invoke this skill when you need the user to act - run a command, paste a secret, click, approve - and whenever they ask how to do something or say they do not know what to do: "step by step", "walk me through it", "what do I do", "what should I do", "I don't understand what to do", "explain what I need to do", in any language. Picking which task comes next is os-whats-next; this skill is for doing the thing in front of you. First earn the ask: try it yourself, find another route, shrink it to the part only they can do. Then one action per step, commands labelled by what they touch, no jargon. Commands are single lines that prompt for any value - typing hidden for secrets - and confirm in plain words. Afterwards verify their step.

日本語の概要は準備中です。原文の説明を表示しています。

kharmanskyi/open-steps1,2842026年10月6日 更新

ALWAYS invoke this skill before anything hard to undo gets agreed to - a contract, a purchase, a migration, a launch, a price change, a reorganisation - and whenever the user asks "what could go wrong", "what are we missing", "poke holes in this", or for a premortem or a red team, in any language. Assumes the decision already failed and works backwards to find out why, in a fresh agent that had no hand in making it. Sweeps nine areas and shows what each produced, including the empty ones. Ends on one verdict: go ahead, go but fix these first, try it small first, think again, do not do this.

日本語の概要は準備中です。原文の説明を表示しています。

kharmanskyi/open-steps1,2842026年10月6日 更新

kharmanskyi のスキルをすべて見る

このスキルの問題を報告する