mcpField
A
android-workmanager-notifications

android-workmanager-notifications

github.com/krutikjainDevelopment1478 installsUpdated 31w ago
Repository

About android-workmanager-notifications

android-workmanager-notifications is a skill for planning and reviewing Android background work and notification delivery. It helps choose between WorkManager, foreground services, and alarms, with attention to retries, duplicate scheduling, permissions, and platform limits.

Key features

  1. 01Selects an execution API for the workload
  2. 02Defines unique work, retries, and cancellation
  3. 03Reviews constraints and expedited work
  4. 04Accounts for notification channels and permission denial
  5. 05Checks reboot, process death, and duplicate scheduling

Use cases

  1. 01Plan a persistent reminder
  2. 02Review a background sync pipeline
  3. 03Check notification behavior when permission is denied

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/krutikjain/android-agent-skills --skill android-workmanager-notifications

After installing, type/android-workmanager-notificationsin Claude Code to confirm it loaded.

Bundled files

References and scripts installed along with SKILL.md.

agents/
openai.yaml
references/
official-links.md
patterns.md
scenarios.md
scripts/
run_examples.sh

Details

From the project README

Android WorkManager Notifications

When To Use

  • Use this skill when the request is about: android workmanager job, background reminder notification android, reliable retry workmanager.
  • Primary outcome: Schedule reliable background work, reminders, and notification delivery with WorkManager and Android execution limits.
  • Read references/patterns.md when you need the API-choice matrix for WorkManager vs foreground service vs alarms.
  • Read references/scenarios.md for unique-work, notification-permission, and reminder-delivery review paths.
  • Handoff skills when the scope expands:
  • android-permissions-activity-results
  • android-performance-observability

Workflow

  1. Choose the right execution API first: WorkManager for deferrable guaranteed work, foreground services for ongoing user-visible work, and exact alarms only when clock-time precision and policy requirements truly demand them.
  2. Make work idempotent and uniquely addressable with explicit names, input data contracts, backoff policy, and cancellation behavior.
  3. Apply constraints, expedited work, and retry semantics only where they match user value and platform limits.
  4. Build notification delivery as part of the contract: channels, importance, tap action, runtime permission handling, and graceful behavior when notifications are denied.
  5. Validate duplicate scheduling, reboot/process death, and degraded states before handing off performance or permission-specific follow-up work.

Guardrails

  • Keep background work idempotent and cancellation-aware.
  • Use unique work or unique periodic work when the feature represents one logical reminder or sync pipeline.
  • Treat expedited work as scarce and user-valuable; do not use it as a generic queue-jumping flag.
  • If a worker uses network constraints, ensure the app declares the permissions required for those constraints on modern Android versions.
  • Keep notification value intact even when the user declines POST_NOTIFICATIONS; the app still needs a visible recovery path.

Anti-Patterns

  • Blocking the main thread with disk or network calls.
  • Expecting WorkManager to deliver exact wall-clock timing like an alarm API.
  • Re-enqueueing duplicate reminder work instead of using unique work policies.
  • Failing to create channels or handle notification permission decline paths.
  • Hiding critical foreground or exact-alarm policy assumptions in comments instead of code and docs.

Review Focus

  • Is WorkManager the right API for this workload?
  • Are unique work policies, constraints, retry, and cancellation explicit?
  • Do reminder and notification paths still make sense if notifications are denied?
  • Are expedited, foreground, or exact-alarm behaviors justified and policy-safe?

Examples

Happy path
  • Scenario: Find background-work and notification hooks before proposing a scheduling change.
  • Command: rg -n "WorkManager|Worker|enqueueUnique|PeriodicWorkRequest|NotificationChannel|POST_NOTIFICATIONS" examples
Edge case
  • Scenario: Review exact-alarm, expedited, or network-constrained work for platform-sensitive behavior.
  • Command: rg -n "setExpedited|setRequiredNetworkType|AlarmManager|ForegroundInfo" examples
Failure recovery
  • Scenario: Disambiguate WorkManager requests from permission prompts and performance work.
  • Command: python3 scripts/eval_triggers.py --skill android-workmanager-notifications

Done Checklist

  • The chosen execution API matches the workload's timing and visibility requirements.
  • Unique work, retry, cancellation, and constraint behavior are explicit.
  • Notification channel and permission behavior are handled as part of the feature contract.
  • Process-death, reboot, and duplicate-scheduling behavior have been considered.

Official References

Comments

Markdown supported