pult-dnya
Skill3 starsChanged 16 days agoApache-2.0View on GitHub
Use on every new task from the owner, when starting, waiting on someone or closing a task, and in the morning, day and evening rhythm: one pinned day board. RU: «пульт дня», новое поручение, что в работе, кто кого ждёт.
In this repository47 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- cardsSkill
- communicationSkill
- copyblockSkill
- delaSkill
- escalate-to-ownerSkill
- fantaziyaSkill
- hodSkill
- import-reviewSkill
- inboxSkill
- inbox-triageSkill
- iniciativaSkill
- konnektorySkill
- librarySkill
- memory-auditSkill
- mila-exportSkill
- MILAGPT-chatSkill
- mila-in-chatSkill
- mila-sessionSkill
- mila-speechSkill
- mila-tasksSkill
- mila-uchitsyaSkill
- mila-updateSkill
- obratnaya-svyazSkill
- oglyadkaSkill
- onboardingSkill
- owner-dmSkill
- packsSkill
- pd-vaultSkill
- plain-wordsSkill
- problemySkill
- raspisanieSkill
- razvilkiSkill
- safe-linkSkill
- self-checkSkill
- shturmanSkill
- sleduyushchij-hodSkill
- sluhSkill
- subscription-limitsSkill
- subscriptionSkill
- talantySkill
- tasks-disciplineSkill
- tasksSkill
- topicsSkill
- utochnenie-zadachSkill
- verifySkill
- vidzhetSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
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 45
Don't write "waiting for you" anywhere except as a question with buttons.
- Distinctive: line 46
Don't start a third task: close one or put it to wait with a question.
- Distinctive: line 47
Don't post the board as a new message every time: edit the pinned one.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Day board
One truth about my day, in one pinned message for the owner. Everything I take, wait for and close goes through the board, so the owner never has to ask "where is it?" and never waits for me while I wait for him.
Script: python3 ~/.claude/skills/pult-dnya/scripts/day_board.py <command>. The day file is ~/work/state/day-<date>.json. Unfinished items carry over to the next day with the same number.
Five rules
- A new task from the owner gets a line at once:
add --text "…" --who owner --why "…"→ «#N queued, after #X, because …». I send that line to the owner in the same minute. I do not drop what I am doing, unless the new task is 🔴 (--red): then it goes straight into progress. - "Waiting for you" exists only as a question asked in the same minute. It is a message with buttons (A/B and my recommendation, checked with plain-words). Only then
wait N --on owner --asked <message link or id>. Without--askedthe script refuses: a passive "waiting" line in a report is not a question. Lesson 22.10: "Mila waits for me, and I wait for her" — that is how a day is lost. - At most two things in progress.
start Nrefuses a third. 🔴 always gets in. - Rhythm:
- Morning:
plan --pool "a; b; c"— the day's plan plus an auto-pool of 3–5 of my own tasks, so the queue is never empty. - Day: after each change,
render→edit_messageof the pinned board. The board is edited, not reposted.pin --chat C --message Mremembers which message is pinned. - Buttons on the board (senior's draft 29.10): one button per «waiting for you» item, at most 6, rows of 1–3, the last one «❓ Ask me the rest» when there are more.
edit_messagechanges the text but not the buttons: when the set of waiting items changes, send a new board withreply+ buttons andpinit; the old one stops being the pinned board. Button format — only{label, choice|url, style}(skillmilagpt-chat); Bot API{text, callback_data}does not work. /statusin Telegram (mila-telegram plugin) shows this board to the owner on demand:renderreturnsempty(no plan today → «Make today's plan» button) andwaiting(items waiting on the owner —--on owner/хозяинor a name fromMILA_OWNER_NAME→ «#N — answer» buttons).rendernever changes the day file.- Evening:
evening→ results with conclusions, what carries over, who waits for whom.
- Morning:
- Closing needs evidence:
done N --evidence <link, file, tool answer> --result "<conclusion in one line>". No evidence — not done.drop N --why …removes a task with a reason.
What the owner sees
📋 Day board · 22.10
▶️ In progress
#7 Discord bridge · owner
⏳ Queue
#8 Day board skill · admin · after #7
🤝 Waiting for an answer
#5 Wallet for Discord · waiting: owner · asked:
✅ Done
#3 macOS tests → rollback fixed · b29e066The board as a widget
Where the Telegram plugin has the widget tools, the day board can be one widget message in the owner's private chat instead of the pinned text: python3 ~/.claude/skills/vidzhet/scripts/widget.py show pult --chat <owner> --bot <my bot> in the morning (then pin it), refresh pult after every change. It shows what is in progress, ONE question that waits for the owner with a single button «Answer», and what is done. The button only asks me to send that question as a usual card — the widget takes no answers. The commands of the board itself (add, start, wait, done) stay the same. Details: the skill vidzhet.
Don'ts
- Don't write "waiting for you" anywhere except as a question with buttons.
- Don't start a third task: close one or put it to wait with a question.
- Don't post the board as a new message every time: edit the pinned one; a new message only for the morning plan, the evening result and a changed set of «waiting» buttons.
- Name the task, not the person: not «call Ivan», but «a call to client X», and only if the chat is internal. The script masks cards (****1234), phones and e-mails anyway, on input and on output.
- A wait older than an hour turns 🔴 on the board: ask again or decide myself.
--askedtakes only a t.me/c/… link or a message number.
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.

