start-the-day
Skill7 starsChanged 49 days agoMITView on GitHub
Morning ADHD-focus ritual. Renames the chat to "Daily DD-MM-YYYY" (Claude desktop app only, skipped elsewhere), optionally checks in about sorting your task inbox and any other capture inboxes (skippable), opens (or creates) today's Daily note, asks whether you want a stale-task review at all (default no) and only runs surface-stale-tasks on a yes, then runs pick-meetings-to-prep, which asks which meetings to prep and runs prep-meeting once per meeting. Trigger: /start-the-day or "start the day" / "kick off my morning".
1 other project uses a version of this.
Adopted by
Most recently taken up by
- vellum-ai/vellum-assistanttheir own version, first seen Jul 8, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- vellum-ai/vellum-assistant★ 1.5ksame name
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.
- CLAUDE.mdCLAUDE.md
- end-taskSkill
- end-the-daySkill
- link-taskSkill
- log-completed-tasksSkill
- pick-meetings-to-prepSkill
- prep-meetingSkill
- search-mailSkill
- setupSkill
- suggest-brag-itemsSkill
- surface-stale-tasksSkill
- update-tasksSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 4 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 99
If it exists, read it - don't clobber anything already written under ## Notes or elsewhere.
- Very specific: line 101
If it doesn't exist, create it using the current ⚙️ Templates/Daily.md as the structure (read the template fresh each time rather than …
- Distinctive: line 106
Do not write a task list into the Daily note - your task manager is the source of truth for what's on today's plate, and mirroring it here …
- Distinctive: line 126
Don't mark anything done in your task manager, don't invent new tasks, don't reorganise unrelated parts of the note.
- Distinctive: line 133
If you're using an external task manager, task tracking lives there, not in the vault - the Daily note doesn't mirror a task list.
- Distinctive: line 137
No em-dashes in anything written into notes - hyphens instead (edit to taste).
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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
Start the Day
Purpose
A focus tool, not a status report. Prioritisation happens in your task
manager, not here - the Daily note doesn't try to mirror a "today's focus"
checkbox list. This skill is an orchestrator only - it does no work
itself beyond opening the Daily note; the actual staleness check and meeting
prep live in their own standalone skills (surface-stale-tasks,
prep-meeting) that this one calls in sequence.
Why split into smaller skills
This used to be one skill doing three jobs. It's been split so that each job is independently useful: you might want a stale-task check mid-afternoon without running the whole morning ritual, or want to prep one extra meeting that gets added to your diary after lunch, without repeating everything else. Splitting also keeps each skill's own job statable in one sentence, which makes each one easier to trust and to edit later without touching the others. See the repo README's "One skill, one job" principle - this pattern is meant to generalise to whatever skills you add later, not just apply here.
Why a separate task manager at all
An earlier version of this skill (and the Tasks plugin Obsidian setup it
was built on) kept everything - notes and tasks - inside the vault, with
Dataview/Tasks queries doing the surfacing. That works, and if it works for
you, keep it: skip this skill's task-manager-integration steps and just use
end-the-day against vault-native #task checkboxes instead (see that
skill's notes).
The author moved task tracking out to a dedicated app (Things3, on macOS)
because a note-taking tool and a task manager are optimising for different
things: a real task manager gives you due-date notifications, recurring
tasks, and capture from anywhere (phone, watch, car) without needing Obsidian
open, none of which a markdown checkbox in a note does well. The tradeoff is
one more system to keep in sync - these skills make that sync asymmetric and
cheap: they almost only read from the task manager. The one exception is
that end-the-day can add a new item to your task inbox when you say yes to
that specific item - it never organises, schedules, tags or completes
anything, so the inbox stays the single place new tasks get sorted, and
there's no risk of Claude quietly diverging from what you actually asked
your task manager to track.
This part is intentionally pluggable. surface-stale-tasks and
prep-meeting describe the pattern (export/query task data, apply
staleness thresholds, filter by tag) rather than a specific tool. Wherever
they say "your task manager," substitute whatever you use - Things3 via
AppleScript (the author's setup; see skills/things3-export/ for example
scripts you'd adapt), Todoist/TickTick via their APIs, Reminders.app, or
anything else that can hand Claude a plain-text or JSON export of open items
with creation/due dates and tags.
Trigger
/start-the-day, or asking to start/kick off your day/morning.
Process
- Rename this chat (optional, Claude desktop app only). Do this first,
before any question. Rename the chat to
Daily DD-MM-YYYY(today's date) using the desktop app's session tools - the app may ask the user to approve the new title. If those tools aren't available (e.g. a terminal session), skip silently. Don't file it into a sidebar group here -end-the-daydoes that at day's end, so today's chat stays easy to reach in the main list all day. This supports a pattern of one long-running "daily" chat for ad hoc questions, with bigger tasks forked into their own chats. - If you keep a task inbox, ask whether to sort it now, skip it, or it's already done. Offer all three - you may have already sorted it before invoking this skill. If you capture tasks anywhere that isn't your task manager's inbox (e.g. voice captures on a phone your task manager has no app for - the author dictates to Gemini in the car, which saves to Google Tasks), include a reminder in the question to clear those into the inbox as part of the same sort. This is just the question - don't pull or list inbox items yourself, that part stays entirely manual in whatever app owns your task inbox. If the user wants to sort it now, wait for them to say they're done before moving on. Don't push back on a skip - the point is this stays optional so it doesn't turn into a chore that gets avoided. Skip this step entirely if you don't use a separate task inbox concept.
- Find or create today's Daily note.
- Path:
📝 Notes/🗓️ Daily notes/YYYY-MM-DD.md(today's date). - If it exists, read it - don't clobber anything already written under
## Notesor elsewhere. - If it doesn't exist, create it using the current
⚙️ Templates/Daily.mdas the structure (read the template fresh each time rather than hardcoding it here - it may change). Compute the date/time directly rather than relying on Templater, since this file is being written directly. - Do not write a task list into the Daily note - your task manager is the source of truth for what's on today's plate, and mirroring it here is duplicated maintenance for no benefit.
- Path:
- Ask whether the user wants to go over stale tasks at all - default is
no. A plain yes/no before anything runs. Many mornings the answer is no,
because the user reviews this themselves while going through their task
manager, so don't run the scan speculatively "just to see". Only on a yes,
run the
surface-stale-tasksskill and present its output as-is - then stop and wait for the user's response before moving on. Never present the stale list and continue to meetings in the same message; that turns the question into decoration. On a no, go straight to the next step without comment. (The author added this gate after the always-on version never once led to an action.) - Run the
pick-meetings-to-prepskill. It asks which meetings the user wants to prep for and runsprep-meetingonce per meeting named - present its output as-is. (It can optionally read a calendar instead - see that skill.) - Don't mark anything done in your task manager, don't invent new tasks, don't reorganise unrelated parts of the note. This is a read-and-surface ritual, not a cleanup or prioritisation step - any actual changes to task state are the user's to make themselves, in the tool that owns that state.
Notes on conventions (see the vault's CLAUDE.md for the full set)
- If you're using an external task manager, task tracking lives there, not in the vault - the Daily note doesn't mirror a task list.
- Flat tags only, never nested, for Obsidian notes. Your task manager's tags are a separate system and don't need to match that convention.
- No em-dashes in anything written into notes - hyphens instead (edit to taste).
- This skill never writes to your task manager. Elsewhere, the only write is
end-the-dayadding a new inbox item on a per-item yes - never organising, completing or rescheduling anything.
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.

