alirezarezvani/claude-skills / productivity/weekly-review/skills/weekly-review
weekly-review
VerifiedUse 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.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 5.7 KB |
| assets/example_weekly_review.md | markdown | 4.1 KB |
| assets/weekly_review_checklist.md | markdown | 2.4 KB |
| references/gtd_weekly_review_canon.md | markdown | 6.2 KB |
| references/open_loop_psychology.md | markdown | 6.0 KB |
| references/review_cadence_design.md | markdown | 6.2 KB |
| scripts/commitment_auditor.py | script | 7.3 KB |
| scripts/open_loop_scanner.py | script | 7.6 KB |
| scripts/weekly_review_gate.py | script | 9.2 KB |
references/open_loop_psychology.md
View on GitHub# The Psychology of Open Loops
> Why unfinished commitments tax attention even when you're not working on them — the research
> behind Allen's "mind like water," and why writing a next action down genuinely quiets the mind.
## The Zeigarnik effect: unfinished tasks stay resident
Bluma Zeigarnik's classic experiments found that interrupted, unfinished tasks are recalled
substantially better than completed ones — the canonical finding is roughly a 2:1 recall
advantage. The interpretation that survived a century of follow-up work: an unfinished task
maintains a state of tension that keeps it active in memory. Completion releases the tension;
interruption preserves it.
For a personal system, the implication is direct: every commitment you've made and not resolved —
the un-replied email, the half-done project, the promise made in a hallway — is an *open loop*
that your memory keeps rehearsing whether or not rehearsing helps. Allen's term for the
accumulated hum of these loops is "psychic RAM": working memory spent tracking commitments instead
of thinking.
## Masicampo & Baumeister: the plan, not the completion, closes the loop
The most important modern refinement for GTD comes from E.J. Masicampo and Roy Baumeister's
"Consider It Done!" studies. Unfulfilled goals were shown to intrude on unrelated tasks — subjects
with an unfinished goal performed worse on reading comprehension and were more distracted by
goal-related words. The striking result: **making a specific plan for the goal eliminated the
intrusions**, even though the goal itself remained unfinished. The mind doesn't demand completion
before it stands down; it demands a *credible plan* held somewhere it trusts.
This is the experimental vindication of the weekly review's core move. You don't have to finish
your projects to stop them from nagging you — you have to give each one a concrete next action in
a system you demonstrably revisit. The "demonstrably revisit" clause is why the review has to
actually happen on cadence: a plan written into a system that never gets read stops being
credible, and the intrusions come back.
## Attention residue: switching without closing costs twice
Sophie Leroy's work on attention residue shows that when people switch from Task A to Task B
without closing A, part of their attention remains allocated to A — and performance on B suffers.
Residue is worst when the prior task is unfinished and low in "closure." Open loops therefore
don't just occupy memory at rest; they degrade the quality of whatever you *are* working on,
because each unclosed thread leaks attention into the current one.
The weekly review is a batch closure operation: one scheduled session in which every dangling
thread gets an explicit disposition (done, next action, waiting-for, someday/maybe, trash), so the
rest of the week can be spent with attention undivided.
## Cognitive load: the system as external memory
John Sweller's cognitive load theory frames working memory as severely capacity-limited (a handful
of elements at a time) while schemas and external supports offload that limit. A trusted external
system is precisely such an offload: it converts "remember to remember" (extraneous load, paid
continuously) into "consult the list" (paid once, at review time). The weekly review is the
maintenance the offload requires — an external memory that drifts from reality re-imports its
contents back into working memory, with interest.
## "Mind like water"
Allen's martial-arts image summarizes the target state: a mind that responds to what's in front of
it with appropriate force and returns to calm — neither over-reacting to a new input nor
under-reacting because it's busy rehearsing forty old ones. The research above cashes the metaphor
out: capture ends rehearsal (Zeigarnik), a next action ends intrusion (Masicampo & Baumeister),
closure ends residue (Leroy), and the external system ends the memory tax (Sweller). The review is
what keeps all four working.
## Practical consequences encoded in this skill
- **Scan, don't recall.** The open-loop scanner walks the actual workspace because self-report of
open loops is exactly what the research says is unreliable — loops intrude, but they don't
enumerate themselves on demand.
- **Every project leaves with one next action.** Not "review the project" — a physical,
visible next step. That's the specific-plan condition from the plan-making studies.
- **Skipping the waiting-for review is a real cost.** Delegated items are open loops too; the
tension doesn't care that someone else is holding the ball.
- **Restart without guilt.** The loops don't punish you for lapsing; they just come back. The
correct response to a lapsed cadence is a shorter review, not a longer apology.
## Sources
1. Bluma Zeigarnik, "Über das Behalten von erledigten und unerledigten Handlungen" ("On Finished
and Unfinished Tasks"), *Psychologische Forschung* 9, 1927 — the original recall-advantage
experiments for interrupted tasks.
2. E.J. Masicampo & Roy F. Baumeister, "Consider It Done! Plan Making Can Eliminate the Cognitive
Effects of Unfulfilled Goals," *Journal of Personality and Social Psychology* 101(4), 2011 —
plan-making eliminates goal intrusions without completion.
3. Sophie Leroy, "Why Is It So Hard to Do My Work? The Challenge of Attention Residue When
Switching Between Work Tasks," *Organizational Behavior and Human Decision Processes* 109(2),
2009 — attention residue from unclosed tasks degrades subsequent performance.
4. John Sweller, "Cognitive Load During Problem Solving: Effects on Learning," *Cognitive Science*
12(2), 1988 (and later cognitive load theory syntheses) — working-memory limits and the case
for external offloading.
5. David Allen, *Getting Things Done* (Penguin, revised edition 2015) — "psychic RAM," "mind like
water," and open loops as the phenomenology the research explains.
6. Roy F. Baumeister & John Tierney, *Willpower: Rediscovering the Greatest Human Strength*
(Penguin, 2011) — the accessible synthesis of the Zeigarnik and plan-making work, including its
explicit connection to GTD.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| 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: 472k | Updated | Claude CodeCodex | Verified | Install |
| 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: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |