Skip to content

alirezarezvani/claude-skills / productivity/weekly-review/skills/weekly-review

weekly-review

Verified

Use when someone wants to run a weekly review, close open loops, audit stalled projects and commitments, get their system back to trusted, restart a lapsed review habit, or says "/cs-weekly-review". Walks David Allen's three-phase loop — GET CLEAR, GET CURRENT, GET CREATIVE — with deterministic scripts that inventory open loops, gate the checklist with named gaps, and score commitment health 0-100.

references/review_cadence_design.md

View on GitHub
# Designing a Review Cadence That Survives Contact With Real Weeks

> Choosing the right horizon for each review, anchoring the habit so it fires without motivation,
> timeboxing so it keeps firing, and restarting after the inevitable lapse.

## Match the review to the horizon

GTD's "horizons of focus" model assigns different altitudes to different cadences. The mistake to
avoid is asking one review to do every altitude's job:

| Horizon | Cadence | What gets reviewed | What it is NOT |
|---|---|---|---|
| Ground — next actions & calendar | Daily (minutes) | What's next *right now*; today's hard landscape | Not a system audit |
| Projects & waiting-for | **Weekly** (60-90 min) | Every list, both calendars, every project's next action | Not goal-setting |
| Areas of responsibility | Monthly-ish | Are whole areas (health, finances, team) getting neglected? | Not task triage |
| Goals, vision, purpose | Quarterly / yearly | Direction: right projects at all? | Not maintenance |

The weekly review is the load-bearing middle: frequent enough that loops can't rot, high enough to
see project drift the daily view can't. Skipping it and hoping the daily check compensates is the
most common cadence failure — daily views show *tasks*, and a system can be current on tasks while
five projects silently stall.

## Anchor the habit; don't schedule the intention

Habit research is unambiguous that repetition after a **stable cue** — not motivation, not
importance — is what makes a behavior automatic. Applied to the review:

- **Fogg's recipe:** After I [existing anchor], I will [first tiny step]. "After Friday's last
  meeting ends, I open the review checklist" beats "do weekly review Friday" — the anchor is an
  event that already happens, and the committed behavior is deliberately tiny (open the checklist,
  not finish the review). Momentum does the rest; celebration seals it.
- **Clear's environment framing:** make it obvious (recurring calendar block with the checklist
  linked), easy (everything in one place — this skill's scripts and checklist), and satisfying
  (the COMPLETE verdict and health score are the visible reward). Habit formation takes weeks to
  months of repetition — expect the review to feel effortful for the first dozen iterations.
- **Protect the block like a meeting with someone important,** because it is: it's the meeting
  where you renegotiate every commitment you've made with yourself.

## Timebox it — the review must be cheap enough to repeat

A review that costs three hours will be skipped the first busy week, and skipped once is the
beginning of lapsed. Working discipline:

- **Target 60-90 minutes; hard-stop at two.** At the stop, gate what's done honestly
  (INCOMPLETE is information, not failure), schedule the remainder, leave.
- **Process, don't do.** The two-minute rule is the *only* execution allowed. Everything else
  becomes a next action. Reviews balloon when they silently turn into work sessions.
- **Let the scripts carry the clerical weight.** Scanning for loops and computing portfolio health
  are mechanical; spend the human minutes on decisions (chase or drop? activate or kill?).

## Failure modes

- **The guilt ritual.** The review becomes a weekly confrontation with everything undone, feels
  bad, and gets avoided. Countermeasures: the someday/maybe list exists precisely so ambitions can
  be parked without shame; the commitment auditor names SOMEDAY-CANDIDATEs so parking is a normal
  verdict, not an admission of defeat.
- **The marathon.** Usually caused by processing debt (inboxes untouched all week) or by doing
  work mid-review. Fix the input side (a 10-minute daily sweep keeps inboxes shallow) and enforce
  process-don't-do.
- **The skimmed review.** All boxes ticked in fifteen minutes, no calendar swept, no project
  touched. Worse than skipping, because it stamps stale lists as fresh. This is exactly what the
  gate's mandatory GET CURRENT rule exists to catch.
- **The floating review.** No anchor, so it happens "when there's time," which is never. Re-anchor
  to a fixed event and shrink the committed first step.
- **Altitude creep.** The weekly review turns into an existential goals session every week and
  becomes exhausting. Goals live at the quarterly horizon; the weekly review only has to make the
  system current.

## Restarting after a lapse

Everyone lapses. The restart protocol that preserves the habit:

1. **No back-processing marathon.** Don't attempt to review the missed weeks; review *now*. Old
   loops are still in the system or will surface in the scan.
2. **Run a shorter pass.** GET CLEAR + the five GET CURRENT steps; explicitly skip GET CREATIVE
   with a reason if energy is short. An honest 80% INCOMPLETE beats a fantasized 100%.
3. **Zero self-punishment.** Missing-then-restarting is what successful habit maintenance actually
   looks like in the data; a single missed instance measurably does not derail formation — quitting
   in response to missing does.
4. **Re-anchor before re-ambitioning.** Fix the cue and shrink the step; don't respond to a lapse
   by designing a grander review.

## Sources

1. David Allen, *Getting Things Done* (Penguin, revised edition 2015) — horizons of focus;
   the weekly review's altitude; the two-minute rule.
2. BJ Fogg, *Tiny Habits: The Small Changes That Change Everything* (Houghton Mifflin Harcourt,
   2020) — anchor → tiny behavior → celebration; behavior = motivation × ability × prompt.
3. James Clear, *Atomic Habits* (Avery, 2018) — cue/craving/response/reward; make it obvious,
   easy, satisfying; "never miss twice."
4. Phillippa Lally et al., "How Are Habits Formed: Modelling Habit Formation in the Real World,"
   *European Journal of Social Psychology* 40(6), 2010 — automaticity takes ~18-254 days; a single
   missed repetition does not materially harm formation.
5. Cal Newport, *Deep Work* (Grand Central, 2016) and weekly/quarterly-planning essays at
   calnewport.com — multi-scale planning; the shutdown ritual as daily loop-closing.
6. Tiago Forte, "The Weekly Review Is an Operating System" (fortelabs.com) — the review as routing
   and maintenance, and why doing work inside it kills it.
7. GTD® methodology guides, gettingthingsdone.com (David Allen Company) — official guidance on
   scheduling, protecting, and recovering the Weekly Review®.
rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first.Stars: 196kUpdated Claude CodeCodexVerifiedInstall