ワンクリックで
weekly-self-improve-loop
Weekly review of memory and blocked items. Extract friction patterns, create skills, and update autonomy baseline.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
メニュー
Weekly review of memory and blocked items. Extract friction patterns, create skills, and update autonomy baseline.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | weekly-self-improve-loop |
| description | Weekly review of memory and blocked items. Extract friction patterns, create skills, and update autonomy baseline. |
Weekly review for continuous improvement.
Without regular review:
# Get last 7 days of memory
$startDate = (Get-Date).AddDays(-7)
$memoryFiles = Get-ChildItem "memory/" -Filter "*.md" |
Where-Object { $_.LastWriteTime -ge $startDate }
# Aggregate metrics
$totalTasks = 0
$completedTasks = 0
$blockedTasks = 0
$patterns = @{}
foreach ($file in $memoryFiles) {
$content = Get-Content $file.FullName -Raw
# Count tasks
$totalTasks += ([regex]::Matches($content, "Task:")).Count
$completedTasks += ([regex]::Matches($content, "Status: complete")).Count
$blockedTasks += ([regex]::Matches($content, "Blocker:")).Count
# Extract patterns
$blockers = [regex]::Matches($content, "Blocker: (.+)")
foreach ($b in $blockers) {
$key = $b.Groups[1].Value
$patterns[$key] = $patterns[$key] + 1
}
}
# Calculate rates
$completionRate = [math]::Round(($completedTasks / $totalTasks) * 100, 1)
$blockerRate = [math]::Round(($blockedTasks / $totalTasks) * 100, 1)
# Find top friction patterns
$topPatterns = $patterns.GetEnumerator() |
Sort-Object Value -Descending |
Select-Object -First 3
foreach ($p in $topPatterns) {
Write-Host "Pattern: $($p.Key) ({$p.Value} occurrences)"
# Create or update skill
$skillName = $p.Key -replace '[^a-z]', '-' -replace '-+', '-'
$skillPath = "skills/local/$skillName-recovery"
if (Test-Path $skillPath) {
Write-Host " Updating existing skill..."
} else {
Write-Host " Creating new skill..."
# Create skill (see memory-to-skill-crystallizer)
}
}
## Weekly Report (YYYY-MM-DD)
**Metrics**:
- Total tasks: X
- Completion rate: Y%
- Blocker rate: Z%
**Top Friction Patterns**:
1. Pattern A (N occurrences)
2. Pattern B (N occurrences)
3. Pattern C (N occurrences)
**New Skills Created**:
- skill-name-1
- skill-name-2
**Next Week Focus**:
- Address pattern A with automated fix
- Review skill effectiveness
| Criteria | Verification |
|---|---|
| Review executed | Report file created |
| Metrics calculated | Completion/blocker rates present |
| Patterns extracted | Top 3 patterns identified |
| Skills created/updated | At least 1 skill actioned |
Use when:
Review weekly. Improve continuously.
Fix OpenClaw connectivity to foreign apps for users in China without resetting existing tasks or tokens. Use when WhatsApp/Telegram/Discord linking fails with WebSocket timeout or handshake errors (for example status=408 Request Time-out, Opening handshake has timed out), when direct outbound access is blocked but VPN/proxy is available, or when OpenClaw can reach local dashboard but channels remain "Not linked".
Coordinate between local skills and discovered ClawHub patterns. Integrates phoenix-loop, agent-audit-trail, HEARTBEAT, and autonomy skills with external best practices.
Publish skills to ClawHub via web dashboard only. No CLI login, no device flow. Reuse existing browser session.
When blocked, report exact blocker + attempts made + minimum user input needed. Enable fast unblock without back-and-forth.
Verify evidence URLs are real and accessible. Check that artifact links resolve to actual content, not placeholders.
Diagnose and fix gateway token mismatches causing 401 errors. Align tokens across config, service, and CLI surfaces.