AgentStack
SKILL verified MIT Self-run

Weekly Review

skill-onurpolat05-alba-weekly-review · by onurpolat05

Weekly performance review - analyze progress, extract patterns, plan next week

No reviews yet
0 installs
11 views
0.0% view→install

Install

$ agentstack add skill-onurpolat05-alba-weekly-review

✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

Are you the author of Weekly Review? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

/weekly-review - Weekly Performance Review

Structured weekly review: gather data, present summary, ask questions, save review.

Process

1. Gather Data (silent - don't show raw data)

Read these files:

  • memory/state/dashboard.md - current priorities and status
  • memory/state/todo.md - task completion
  • memory/knowledge/learnings.md - this week's entries (by date)
  • memory/knowledge/errors.md - this week's errors (by date)
  • memory/daily/ - scan for this week's session logs (Mon-Sun)

Extract:

  • Tasks completed: count [x] items from dailies + todo
  • Tasks added: count new [ ] items
  • Learnings recorded this week
  • Errors encountered and resolved
  • Open blockers from dashboard

2. Present Draft

Show a concise draft:

Week Review: YYYY-Www (Mon DD - Sun DD)

Completed: X tasks
- [top 3-5 completed items]

Learnings: Y new
- [list if any]

Errors: Z encountered, W resolved
- [list if any]

Open Blockers:
- [from dashboard, if any]

3. Ask 3 Questions (one at a time)

Q1: "Does this look right? Anything to add or fix?" Q2: "What went well this week?" Q3: "Top 3 priorities for next week?"

Keep questions short. Accept brief answers.

4. Save Review

Write to memory/state/weekly-reviews/YYYY-Www.md:

# Week YYYY-Www

## Summary
- Days active: [count from daily logs]
- Tasks completed: [count]
- New learnings: [count]

## Completed
- [Items from draft + Q1 corrections]

## What Went Well
- [From Q2]

## Learnings
- [This week's entries]

## Blockers
- [Open items]

## Next Week Priorities
1. [From Q3]
2. [From Q3]
3. [From Q3]

5. Update Dashboard

  • Replace/update priority section in memory/state/dashboard.md with next week's priorities from Q3
  • Keep existing project info, only update priorities

6. Archive Check

If memory/state/weekly-reviews/ has more than 4 review files:

  • Suggest: "You have [N] weekly reviews. Archive older ones? (I'll move them to memory/state/weekly-reviews/archive/)"
  • Only archive if user agrees

Output

End with a 3-line confirmation:

Review saved: memory/state/weekly-reviews/YYYY-Www.md
Dashboard updated with next week priorities.
[If archive suggested: "Consider archiving old reviews with /extend"]

Rules

  • Keep the review file under 60 lines
  • Don't generate fake data - if no daily logs exist, say "No session logs found for this week"
  • Use ISO week numbers (W01-W52)
  • Create memory/state/weekly-reviews/ directory if it doesn't exist
  • Match the user's language from CLAUDE.md
  • This is a reflection tool, not a performance evaluation - keep tone neutral and constructive

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet — be the first.

Versions

  • v0.1.0 Imported from the upstream source.