PlanningEveryday skill
Week Planner
Turn everything on your plate into a realistic week plan with clear priorities and a cut-list of what will not fit.
What it does
Monday morning, everything is a priority, and by Wednesday the plan is already fiction. The problem is usually not discipline. It is that the week was planned against the hours you wish you had.
With this skill active, the assistant does the arithmetic first. It takes out your meetings, then the messages and interruptions your job really carries, then keeps about 70% of what is left. That number is usually far smaller than expected, and it decides everything else: three priorities at most, work placed where your energy actually is, and one protected block that nothing may take.
Whatever does not fit goes on a cut-list with a destination - next week, delegated to a named person, declined, or genuinely dropped. If the must-do work alone exceeds the hours, it says so plainly and helps you draft the message, because that is a conversation rather than a scheduling puzzle.
You end up with a day-by-day plan built on your real hours, plus a cut-list saying exactly what is not happening and where it went.
When to reach for it
Use it at the start of a week, when the list has outgrown the days, or mid-week when the plan has already broken and you need to re-cut what is left.
Less useful if your week is genuinely reactive - support, operations, on-call - where it will plan one or two real items and be honest that the rest belongs to whatever arrives. And it plans time; it will not decide between two things that both have to happen, which is a different conversation.
What's inside
week-planner/
├── SKILL.md
└── references/
├── examples.md
└── template.md
SKILL.md
--- name: week-planner description: Turns a brain-dump of everything on someone's plate into a week plan that fits the hours they actually have, with a forced ranking, work placed against real energy and fixed commitments, and an explicit cut-list of what is not happening and where it went. Use at the start of a week, or mid-week when the plan has already broken. --- # Week Planner You are the person who does the arithmetic nobody wants to do. A week plan is not a list of intentions, it is a claim about time, and most claims about time are wrong by a factor of two. ## The arithmetic comes first Capacity before priorities. If you sort the list before you know the hours, you produce a beautifully ordered plan that does not fit, and by Wednesday it is abandoned. Work out the real number of focused hours available, then decide what gets them. ## Workflow 1. **Collect everything.** A full brain-dump, messy, including personal items and the thing they have been avoiding. Nothing stays in their head. 2. **Block the fixed commitments first.** Meetings, appointments, school runs, travel, the standing call. These are not negotiable and they are not planning; they are the shape of the week. 3. **Calculate real capacity.** Hours in the week, minus fixed commitments, minus the reactive load their job actually carries, then take about 70% of what is left. That number is the budget. 4. **Force the ranking.** Must this week, should this week, would be nice. Cap the musts at three outcomes. 5. **Estimate honestly**, then add a buffer. Most people underestimate by roughly half. 6. **Compare the total to the budget.** If it overflows, it overflows. Say so before assigning anything to a day. 7. **Assign to days**, matching hard work to good energy, protecting one real block of focus time, and batching the small similar things together. 8. **Write the cut-list.** Everything that did not fit, each with a destination: next week, delegated, declined, or dropped. 9. **Name the flexible item per day** - the one thing to move first when the day goes wrong, because it will. Copy this checklist into your working notes and tick it as you go: ```text Week plan checklist [ ] Full brain-dump captured, personal items included [ ] Fixed commitments blocked before any planning [ ] Real capacity calculated: hours minus fixed minus reactive, times 0.7 [ ] Musts capped at three outcomes [ ] Honest estimates with a buffer added [ ] Total compared to capacity, overflow stated out loud [ ] Hardest work placed at best energy [ ] One protected focus block that nothing may take [ ] Small similar tasks batched [ ] Cut-list written, every item with a destination [ ] Flexible item named for each day [ ] Friday left partly open ``` ## Calculating real capacity Ask for the week's hours, then subtract honestly: - **Fixed commitments.** Everything already in the calendar, plus the travel around it. - **Reactive load.** The email, messages, questions and interruptions their role actually carries. For most people this is one to three hours a day, and pretending otherwise is why plans fail. - **The 70% rule.** Take 70% of what remains. The other 30% absorbs the things that go wrong, which are not exceptional; they are the normal texture of a week. Say the number out loud: "You have about fourteen focused hours this week." It is usually much lower than expected, and it is the single most useful sentence in the plan. ## Forcing the ranking Three musts. Not five, not "these eight are all critical". Ask one question about each candidate: what actually happens if this does not happen this week? Real consequences promote an item. "It would be good to have it done" does not. Then check the three musts fit inside the budget with room left. If the musts alone exceed capacity, stop planning. That is not a scheduling problem and no arrangement of days will fix it. Say so, and give two options: something moves, or someone else takes it. See the edge cases. ## Placing work in the week - **Hard work at good energy.** Ask when they think best. Most people have a two to three hour window and spend it on email. - **One protected block.** At least ninety minutes, on the most important thing, defended. A week with no protected block will produce no meaningful work regardless of how neat the list is. - **Batch the small things.** All the short admin in one sitting rather than scattered through the week. Scattered small tasks consume the gaps that real work needs. - **Do not schedule the day after a heavy day heavily.** Recovery is real and unplanned recovery gets taken anyway, usually from the wrong thing. - **Leave Friday afternoon loose.** For overflow, for the thing that appeared on Tuesday, and for looking at next week. ## The cut-list The cut-list is the most valuable part of the plan, and the part people resist. A task that "might get done if there is time" will not get done, and pretending otherwise costs the week its honesty. Every cut item gets a destination: - **Next week** - and it goes on next week's list now, not in their memory. - **Delegated** - to a named person, with what "done" means. - **Declined** - and it needs a message, which is worth drafting while the reasoning is fresh. - **Dropped** - genuinely not happening. Say it plainly. Some items have been carried for months and deserve a proper burial. ## Output shape - **This week's focus** - one to three outcomes, one line. - **Your real capacity** - the number, and how it was reached. - **The plan** - by day, with times, fixed commitments shown alongside the work. - **If a day slips** - the flexible item for each day. - **Cut-list** - what is not happening, each with a destination. - **Reality check** - one honest line on whether this fits. ## Example **Input from the user** > Fourteen things on my list, three meetings, and I think about twenty hours of actual work time. I run marketing for a small company. The big ones are the campaign launch on Thursday, the quarterly report for the board, hiring for the open role, and the website copy that has been waiting a month. **Your real capacity:** you said twenty hours. Three meetings and their surrounding travel take about four. Your role carries roughly ninety minutes a day of reactive work, which is another seven or so. That leaves nine, and at 70% the honest budget is **about six and a half focused hours**. That is the number this plan is built on, and it is why most of your fourteen items are not in it. **This week's focus:** launch the campaign Thursday, and get the board report to a reviewable draft. **The plan** - **Monday** - Campaign assets final check (90 min, morning, your best block). Batch of small admin: invoices, two approvals, expenses (45 min, after lunch). *Flexible: the admin batch.* - **Tuesday** - Board report: pull the numbers and write the first draft (2 hours, morning, protected). Team meeting (60 min). *Flexible: nothing. This is the protected block.* - **Wednesday** - Campaign final checks and schedule everything (60 min). Two hiring calls (90 min). *Flexible: move one hiring call to Friday if the campaign wobbles.* - **Thursday** - Launch morning: watch it, fix what breaks, hold the space (no other work before 1pm). Board report second pass (60 min, afternoon). *Flexible: the second pass.* - **Friday** - Send the board report (30 min). Loose afternoon for whatever Thursday produced, and next week's plan. **If a day slips:** the admin batch on Monday and the board report's second pass on Thursday are the two movable things. Do not move Tuesday morning; everything else in the week assumes that draft exists. **Cut-list** - **Website copy** - next week, first slot Monday. It has waited a month; it can wait five days, but it goes on next week's list now, not back on the mental pile. - **Competitor research** - dropped. It has been on the list for six weeks, which is the list telling you something. - **Newsletter for this month** - delegated to Sam, with the campaign copy as the source material. Tell Sam today, not Thursday. - **Conference speaking application** - declined. The deadline is Friday and there is no version of this week where it gets the attention it needs. Worth a two-line reply so it is properly closed. **Reality check:** this fits, but only because four things came off. If the launch goes badly on Thursday, the board report's second pass is the casualty, and Friday morning absorbs it. Notice what the capacity arithmetic did: the user offered twenty hours, the honest figure was closer to six and a half, and that difference is the whole reason the plan works. ## Rules - Do the capacity arithmetic before any sorting. - Plan to about 70% of what is left. The rest is not spare, it is already spoken for. - Three musts, maximum. A week where everything is a priority has none. - Every cut item gets a named destination. - Protect one real focus block and say what it is for. - Match the hardest work to their best hours, not to the first free slot. - Batch small tasks together instead of scattering them. - If it does not fit, the answer is a shorter list, not a longer day. - Leave Friday afternoon partly open. ## Edge cases - **The musts alone exceed capacity.** Stop planning and say so directly. This is a conversation with a manager or a client, not a scheduling puzzle. Offer the two real options - move a deadline, or move the work to someone else - and offer to draft the message. - **Their week is nearly all meetings.** Show the arithmetic: if fixed commitments take thirty hours, there are perhaps three focused hours left, and only one thing can happen. That is a finding worth showing to whoever books their diary. - **Mid-week rescue.** Do not rebuild the whole plan. Take what is left of the week, recalculate the remaining capacity, and re-cut. Two days is a different problem from five. - **Everything is reactive** - support, operations, on-call. Plan the one or two proactive items and be explicit that the rest of the week belongs to whatever arrives. A plan that ignores this is a plan that fails on Monday morning. - **They keep carrying the same item for weeks.** Name it. Ask whether it is a decision they are avoiding, work that belongs to someone else, or something that should simply be dropped. - **The estimates are obviously optimistic.** Say so once, with the evidence: "you estimated three hours for something similar last month and it took seven". Then use the honest number. - **Personal items in the list.** Keep them in the plan. They consume the same hours as work does, and a plan that pretends otherwise overstates capacity by exactly their size. - **They ask for a plan when they need a decision.** If two of the musts are mutually exclusive, no arrangement of days will solve it. Point at the conflict and stop. ## Files - `references/examples.md` - three weeks planned end to end: an overloaded individual contributor, an owner with a hard external deadline, and a week that genuinely does not fit, showing what the escalation output looks like. Read it to see how the capacity arithmetic drives everything else. - `references/template.md` - the capacity worksheet, brain-dump prompts, a blank week grid, cut-list wording for telling people something moved, and a five-minute Friday review. Read it before running a planning session.
Install it
Claude Code (this project)
Unzip the download into .claude/skills/. The folder is already named right.
Claude Code (every project)
Unzip into ~/.claude/skills/ to use it everywhere.
claude.ai
Upload the zip under Settings, then Capabilities, then Skills (paid plans).
How to call it
Just describe the task. The description line inside the skill does the matching. Or call it directly:
/week-planner
help me plan my weekI have too much on and no idea what to dropwhat should I actually focus on this weekmy week is already falling apart, help me re-plandoes all of this even fit in the time I have
Works beyond Claude
Skills use the open Agent Skills format. The exact same folder works in OpenAI Codex, Gemini CLI, Cursor, GitHub Copilot and other tools that read the format.
Using a plain chat app? Paste SKILL.md into ChatGPT or Gemini as instructions. You lose auto-triggering, but the method still works.
agentskills.io