Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Create and manage scheduled subscription tasks. Use when the user wants to set up recurring reminders, periodic reports, scheduled checks, or any automated tasks that run on a schedule. Supports cron expressions, fixed intervals, and one-time executions.
.claude/skills/wecode-ai-subscription-manager/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✓→✗ | ▼ Worse | 138% | 0% |
| case-04 | ✓→✓ | = Same ✓ | 147% | 0% |
| case-05 | ✓→✓ | = Same ✓ | 158% | 0% |
| case-06 | ✓→✓ | = Same ✓ | 427% | 0% |
| case-01 | ✗→✗ | = Same ✗ | 2% | 0% |
You now have access to subscription management tools. Use them to create scheduled, recurring, or periodic tasks for the user.
ALWAYS call this when the user mentions scheduling intent.
Generates a preview of the subscription configuration without creating it. The preview is displayed as an interactive block card in the UI with "Confirm" and "Cancel" buttons.
Key Parameters:
display_name (string): Human-readable task nametrigger_type (string): "cron", "interval", or "one_time"prompt_template (string): The prompt to execute on each runcron_expression (string): For cron type, e.g., "0 9 * * *" (daily at 9am)interval_value + interval_unit: For interval type, e.g., 30 + "minutes"execute_at (string): For one_time type, ISO format datetimepreserve_history (boolean): Whether to keep conversation context across runsexpiration_type + expiration_fixed_date/expiration_duration_days: Optional expirationWorkflow:
preview_subscription with appropriate parameterscreate_subscription with the same parameters to create the subscriptionIMPORTANT:
create_subscription with the same parameterspreview_subscription, you must wait for user to click the "Confirm" button in the preview card or explicitly send a confirmation message. Do NOT auto-create the subscription after preview.create_subscription when user explicitly confirms via message - If user sends "确认" or "创建" after preview, then call create_subscription with the same parameters to actually create the subscription. If user clicks the Confirm button in UI, frontend handles creation automatically - you do NOT need to call create_subscription.preserve_history: true for tasks needing context continuity (daily reports, monitoring)preserve_history: false for independent tasks (reminders, checks)preview_subscription(
display_name="Daily Morning Water Reminder",
trigger_type="cron",
cron_expression="0 9 * * *",
prompt_template="Remind me to drink a glass of water to start the day healthy!",
preserve_history=false
)preview_subscription(
display_name="Weekly Project Summary",
trigger_type="cron",
cron_expression="0 18 * * 5",
prompt_template="Generate a weekly summary of project progress, blockers, and next week's priorities.",
preserve_history=true,
history_message_count=20
)preview_subscription(
display_name="Server Status Monitor",
trigger_type="interval",
interval_value=30,
interval_unit="minutes",
prompt_template="Check server CPU and memory usage. Alert if CPU > 80% or memory > 90%.",
preserve_history=false
)preview_subscription(
display_name="Deployment Reminder",
trigger_type="one_time",
execute_at="2025-04-15T14:00:00",
prompt_template="Remind me to deploy the new feature to production.",
preserve_history=false
)preview_subscription(
display_name="Trial Monitoring",
trigger_type="interval",
interval_value=1,
interval_unit="days",
prompt_template="Check trial account status and send daily summary.",
expiration_type="duration_days",
expiration_duration_days=30
)preview_subscription returns:
json{ "__silent_exit__": true, "reason": "subscription_preview block displayed; waiting for user confirmation", "preview_id": "preview_abc123", "execution_id": "exec_xyz789" }
The frontend will display an interactive preview block card with Confirm/Cancel buttons.
User confirmation handling:
create_subscription with the same parameters to create the subscription| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 9,750 | 6,756 | -31% | 1 | 1 | 0% | 1,735 | 1,778 | +2% | 0 | 0 | — |
case-02 | fail→fail | 8,264 | 7,992 | -3% | 1 | 1 | 0% | 1,226 | 1,932 | +58% | 0 | 0 | — |
case-03 | fail→fail | 4,370 | 16,894 | +287% | 1 | 1 | 0% | 590 | 2,864 | +385% | 0 | 0 | — |
case-04 | pass→pass | 5,846 | 4,375 | -25% | 1 | 1 | 0% | 836 | 2,068 | +147% | 0 | 0 | — |
case-05 | pass→pass | 3,930 | 2,899 | -26% | 1 | 1 | 0% | 739 | 1,907 | +158% | 0 | 0 | — |
case-06 | pass→pass | 2,958 | 3,928 | +33% | 1 | 1 | 0% | 374 | 1,971 | +427% | 0 | 0 | — |
case-07 | fail→fail | 2,669 | 6,070 | +127% | 1 | 1 | 0% | 397 | 1,790 | +351% | 0 | 0 | — |
case-08 | pass→fail | 6,706 | 17,342 | +159% | 1 | 1 | 0% | 1,088 | 2,588 | +138% | 0 | 0 | — |
case-09 | fail→fail | 8,285 | 6,937 | -16% | 1 | 1 | 0% | 1,258 | 1,833 | +46% | 0 | 0 | — |
case-10 | fail→fail | 7,972 | 6,161 | -23% | 1 | 1 | 0% | 1,369 | 1,853 | +35% | 0 | 0 | — |
case-11 | fail→fail | 5,827 | 6,885 | +18% | 1 | 1 | 0% | 928 | 1,887 | +103% | 0 | 0 | — |
case-12 | fail→fail | 11,209 | 8,310 | -26% | 1 | 1 | 0% | 1,901 | 1,933 | +2% | 0 | 0 | — |
case-13 | fail→fail | 4,871 | 14,448 | +197% | 1 | 1 | 0% | 734 | 2,341 | +219% | 0 | 0 | — |
case-14 | fail→fail | 9,278 | 6,896 | -26% | 1 | 1 | 0% | 1,495 | 1,855 | +24% | 0 | 0 | — |
case-15 | fail→fail | 7,819 | 7,947 | +2% | 1 | 1 | 0% | 1,279 | 1,863 | +46% | 0 | 0 | — |
case-16 | fail→fail | 11,393 | 8,411 | -26% | 1 | 1 | 0% | 1,845 | 2,079 | +13% | 0 | 0 | — |
case-17 | fail→fail | 2,308 | 5,931 | +157% | 1 | 1 | 0% | 367 | 1,824 | +397% | 0 | 0 | — |
case-18 | fail→fail | 10,103 | 7,288 | -28% | 1 | 1 | 0% | 1,898 | 1,978 | +4% | 0 | 0 | — |
case-19 | fail→fail | 10,362 | 7,124 | -31% | 1 | 1 | 0% | 1,782 | 1,992 | +12% | 0 | 0 | — |
case-20 | fail→fail | 10,489 | 7,450 | -29% | 1 | 1 | 0% | 1,785 | 1,846 | +3% | 0 | 0 | — |
case-21 | fail→fail | 6,825 | 7,281 | +7% | 1 | 1 | 0% | 1,016 | 1,888 | +86% | 0 | 0 | — |
case-22 | fail→fail | 4,156 | 17,360 | +318% | 1 | 1 | 0% | 570 | 2,585 | +354% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 3 counted toward the lift figure. The other 19 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of -5 percentage points is the difference between those two pass rates over the 3 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
Other measured skills in the registry, with their headline benchmark lift.