agentleFS
wernerbattwernerbatt/obsidian-gtd-cli/.claude/skills/review/SKILL.md

review

Skill0 starsChanged 7 months agoMITView on GitHub

Conduct weekly GTD review

192 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use close-spec, design-with-images and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • dzhngSkill

    close-spec

    When a spec under specs/<feature>/ is done shipping (a write-spec build, or any planned task that produced a spec), archive it to specs/done/ and rewrite it from a build-plan into a durable rationale — the why, the principles, the invariants — pointing back to the real code for the how. Use when implementation has landed and the plan no longer matches what shipped, or the user says a feature/spec is finished. Pairs with [write-spec](../write-spec/SKILL.md) (the plan this closes) and [review](../review/SKILL.md) (run before closing).

    Also in 5 projects that use this

  • dzhngSkill

    design-with-images

    Explore UI and visual design with image generation, then iterate the real implementation against the user-selected concept. Use when brainstorming visual directions, offering mockup variants, or implementing an approved generated design; do not stop at the mockups or the first code approximation.

    Also in 5 projects that use this

  • LambenthanSkill

    daily-arxiv

    每日从 arXiv 拉取新论文,过滤相关性,auto-ingest 高优先级论文,检测 SOTA 更新

    Also in 5 projects that use this

  • LambenthanSkill

    exp-eval

    实验判决门:Review LLM 独立评判实验结果 → 4 种判决路径 → 自动更新 claims confidence、ideas status、graph edges

    Also in 5 projects that use this

  • +3 more from dzhng
  • +5 more from Lambenthan

The file

Show its rules and text

We split this skill into the rules it gives an agent

25 rules. 0 are common, 20 are distinctive, and 5 are very specific to this project.

  • 0

    Common

    Most of the 157 edited versions we read keep these.

  • 20

    Distinctive

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    IMPORTANT: Always use the system clock (date, new Date()) for today's date — never rely on the system prompt date, which may be stale.

    line 15
  2. Distinctive:

    Before starting, check for the most recent weekly review summary:

    line 34
  3. Distinctive:

    Read it and surface any "Still On The Radar" items from last time.

    line 40
  4. Very specific:

    When running the review interactively, read the calendars section from config.yaml to get the account email and calendar IDs, then pull …

    line 71
  5. Distinctive:

    Follow up on pending responses

    line 83
  6. Distinctive:

    Look for "Automated task update" commits.

    line 118
  7. Distinctive:

    If you have many items to process:

    line 136
  8. Very specific:

    Use the /clarify skill to process inbox items

    line 138
  9. Distinctive:

    If tasks are consistently overdue:

    line 141
  10. Distinctive:

    Consider if tasks are still relevant

    line 143
  11. Distinctive:

    If projects have no next actions:

    line 148
  12. Distinctive:

    If YES: Define the very next action

    line 150
  13. Distinctive:

    If NO: Move to Someday/Maybe or archive

    line 151
  14. Distinctive:

    If context lists are overwhelming:

    line 154
  15. Distinctive:

    Consider if you're over-committed

    line 156
  16. Distinctive:

    Distraction-free environment: - No interruptions

    line 171
  17. Distinctive:

    Review everything: - Don't skip sections

    line 172
  18. Distinctive:

    Update context lists: - Keep them current

    line 173
  19. Distinctive:

    Look ahead: - Check calendar and upcoming commitments

    line 175
  20. Distinctive:

    Trust your system: - If it's not in GTD, it doesn't exist

    line 176
  21. Distinctive:

    Start with just Get Clear (inbox to zero)

    line 244
  22. Distinctive:

    Use timer (Pomodoro) to stay focused

    line 246
  23. Very specific:

    Use the /clarify skill for efficient processing

    line 250
  24. Distinctive:

    Keep project list smaller (5-10 active max)

    line 255
  25. Very specific:

    Use --stale-projects during review to catch issues early

    line 288

“Common” is measured: kept by at least 6 in 10 of the 157 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

GTD Review Skill

Facilitate the weekly review process to keep your GTD system current and maintain perspective.

Vault Operations

Use the /obsidian skill for all vault reads and writes. Load it before running any commands: → .claude/skills/obsidian/SKILL.md

IMPORTANT: Always use the system clock (date, new Date()) for today's date — never rely on the system prompt date, which may be stale.

Key queries for the review:

  • Inbox count: Dataview eval inbox query → .length
  • Context distribution: Dataview eval context-counts query
  • Overdue tasks: Dataview eval overdue query
  • Stale projects: Dataview eval stale-projects query
  • Completed this week: Dataview eval completed-tasks query (last 7 days)
  • Someday/maybe: Dataview eval someday query

The Weekly Review

The weekly review is the most critical GTD practice. It's your opportunity to:

  1. Get Clear - Process inbox to zero
  2. Get Current - Review all your commitments
  3. Get Creative - Think about new possibilities

Step 0: Read Last Week's Summary

Before starting, check for the most recent weekly review summary:

# Read vault_path from config.yaml
VAULT_PATH=$(python3 -c "import yaml; print(yaml.safe_load(open('config.yaml'))['vault_path'])")
ls -t "$VAULT_PATH/GTD/Weekly Review "*Summary.md | head -1

Read it and surface any "Still On The Radar" items from last time. These become the first things to check in the current review — were they addressed, or do they carry forward?

Review Report

Generate the review report by running the /obsidian skill queries and assembling the results. The report includes:

1. Get Clear

  • Inbox count - How many items need processing
  • Checklist for clearing all inboxes (digital and physical)
  • Reminder to review daily notes

2. Get Current

Context Review:

  • Count of active tasks by context (@deep, @quick, @batch, @work, @home, etc.)
  • Total active task count

Overdue Tasks:

  • Tasks with scheduled or due dates before today
  • File location and line numbers for quick access

Projects Review:

  • Projects without next actions (stale projects)
  • Checklist for reviewing all active projects

Calendar Review:

  • Past week review
  • Upcoming week preview
  • Future commitments

Live Calendar Data (gccli): When running the review interactively, read the calendars section from config.yaml to get the account email and calendar IDs, then pull real calendar events:

# Primary calendar — past week
gccli <account> events primary --from YYYY-MM-DDTHH:MM:SSZ --to YYYY-MM-DDTHH:MM:SSZ

# Additional calendars (loop through calendars.additional in config)
gccli <account> events <calendar-id> --from ... --to ...

Note: Date format must be YYYY-MM-DDTHH:MM:SSZ (UTC with Z suffix). Plain YYYY-MM-DD returns Bad Request.

Waiting For:

  • Review @waiting items
  • Follow up on pending responses

3. Get Creative

  • Review Someday/Maybe list
  • New projects or ideas
  • Review ponderables

4. Completed This Week

  • Tasks completed in the last 7 days
  • Separate genuine wins (non-recurring tasks completed) from recurring maintenance (tasks with 🔁) — highlight the wins, summarise the recurring as a group

To save the review report, use the /obsidian skill's create command to write to GTD/Weekly Review YYYY-MM-DD Summary.md.

Weekly Review Checklist

Located at: GTD/Checklists/Weekly Review Checklist.md

Review Frequency

Weekly Reviews:

  • Schedule a recurring time (e.g., Friday afternoon, Sunday evening)
  • Block 30-60 minutes
  • Make it non-negotiable

Quick Daily Reviews:

  • Morning: Review today's scheduled tasks
  • Evening: Process inbox, update tomorrow's list

Infrastructure Check

During the review, verify that the daily overdue script ran recently:

VAULT_PATH=$(python3 -c "import yaml; print(yaml.safe_load(open('config.yaml'))['vault_path'])")
cd "$VAULT_PATH" && git log --oneline -3

Look for "Automated task update" commits. If the last one is >2 days old, the scheduled task for daily scripts may need attention.

Context & Priority Distribution

As part of Get Current, use the /obsidian skill's context distribution Dataview eval query to generate counts for all active (non-⏬) tasks by context tag. Track these week-over-week to spot trends (growing lists, priority inflation, etc.).

Weekly Review Summary File

At the end of the review, save a summary to GTD/Weekly Review YYYY-MM-DD Summary.md. Follow the format of prior summaries in that folder. Include:

  • Metrics (inbox, overdue, stale projects, active tasks, completed)
  • Context and priority distribution
  • Calendar highlights (past and upcoming week)
  • Actions taken (purge results, fixes, infrastructure changes)
  • Items still on the radar for next time

Common Review Findings

High Inbox Count

If you have many items to process:

  • Block time for processing
  • Use the /clarify skill to process inbox items

Many Overdue Tasks

If tasks are consistently overdue:

  • Re-evaluate scheduled dates
  • Consider if tasks are still relevant
  • Move to Someday/Maybe if not urgent
  • Break down large tasks

Stale Projects

If projects have no next actions:

  • Decide if project is still active
  • If YES: Define the very next action
  • If NO: Move to Someday/Maybe or archive

Too Many Active Tasks

If context lists are overwhelming:

  • Be ruthless with Someday/Maybe moves
  • Consider if you're over-committed
  • Focus on highest-impact items

Integration with Other Skills

The weekly review works best combined with:

  1. /obsidian — All vault queries and edits
  2. /clarify — Process inbox to zero
  3. /organize — Retag and reschedule tasks
  4. /project — Fix stale projects, create new ones
  5. /purge — Cull bloated context lists

Best Practices

  1. Same time each week - Make it a ritual
  2. Distraction-free environment - No interruptions
  3. Review everything - Don't skip sections
  4. Update context lists - Keep them current
  5. Reflect on wins - Review completed tasks
  6. Look ahead - Check calendar and upcoming commitments
  7. Trust your system - If it's not in GTD, it doesn't exist

Metrics to Track

The weekly review provides these metrics:

  • Inbox count - Trending toward zero?
  • Total active tasks - Manageable or overwhelming?
  • Overdue tasks - Improving or getting worse?
  • Stale projects - Are projects moving forward?
  • Completed tasks - What did you accomplish?

Track these over time to identify patterns and improve your system.

Example Weekly Review Session

[Friday 5:00 PM - Weekly Review]

1. Run review report:
   [Run /obsidian skill queries for inbox, overdue, stale, completed]

   Output:
   - Inbox: 12 items
   - Active tasks: 47
   - Overdue: 5 tasks
   - Stale projects: 2
   - Completed: 23 tasks

2. Process inbox to zero:
   [Use /clarify skill]
   [Process all 12 items...]

3. Address overdue tasks:
   [Review each overdue task, reschedule or complete]

4. Fix stale projects:
   [Add next actions to "Website Redesign" and "Garden Planning"]

5. Review context lists:
   [@deep: 8 tasks - looks good]
   [@quick: 5 tasks - can knock out today]
   [@batch: 3 tasks - schedule batch time]
   [@work: 8 tasks - manageable]
   [@home: 12 tasks - schedule for weekend]

6. Review Someday/Maybe (lowest priority tasks):
   [Found 2 items to activate, added to @ai list]

7. Final check:
   [Re-run /obsidian queries]

   Output:
   - Inbox: 0 items ✓
   - Active tasks: 51
   - Overdue: 0 tasks ✓
   - Stale projects: 0 ✓
   - Completed: 23 tasks

8. Week ahead:
   [Check calendar, schedule time blocks for priority tasks]

Total time: 45 minutes
Result: Clean, current, trusted system!

Troubleshooting

Review feels overwhelming:

  • Start with just Get Clear (inbox to zero)
  • Gradually add other sections
  • Use timer (Pomodoro) to stay focused

Can't reach inbox zero:

  • Block dedicated processing time
  • Use the /clarify skill for efficient processing
  • Be ruthless with trash/someday decisions

Projects always stale:

  • Review projects more frequently (bi-weekly)
  • Keep project list smaller (5-10 active max)
  • Move inactive projects to Someday/Maybe

Too many active tasks:

  • Review and cull context lists
  • Move nice-to-haves to Someday/Maybe
  • Accept you can't do everything

Related Skills

  • /obsidian — All vault read/write operations
  • /clarify — Inbox processing
  • /organize — Context tagging and project creation
  • /project — Project lifecycle management
  • /purge — List reduction

Resources

  • Weekly Review Checklist: GTD/Checklists/Weekly Review Checklist.md
  • Projects List: GTD/Projects List.md
  • Dashboard: GTD/Dashboard.md

Closing the Review

Always end the review by saving a weekly summary. This is the last step — do not consider the review complete without it.

Save to GTD/Weekly Review YYYY-MM-DD Summary.md following the format of prior summaries. The agent should generate this automatically at the end of the review session, even if the user doesn't ask.

Tips

  • Save weekly reviews with dates for historical tracking
  • Compare metrics week-over-week to see trends
  • Celebrate completed tasks before planning ahead
  • Use --stale-projects during review to catch issues early
  • The weekly review is when you sharpen the saw

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.