发现最适合你需求的 AI 技能
Add educational comments to the file specified, or prompt asking for file to comment if one is not provided.
Use when the user asks to control, operate, or automate an Android phone via ADB — tapping, swiping, typing, launching apps, or any UI interaction on a connected device
Adaptive code review that routes to haiku/sonnet/opus based on diff complexity signals. Use instead of requesting-code-review for cost-efficient reviews.
Hierarchical memory management for AI agents across sessions. Maintains three layers — daily notes (raw logs), active context (working memory), and long-term memory (curated knowledge) — with automatic distillation from raw notes to permanent memory. Use when setting up persistent memory for an agent workspace, when an agent needs to remember context across sessions or compaction boundaries, when organizing what to remember vs. forget, or when consolidating scattered notes into structured long-t
Create adaptive learning flashcard apps from course materials (URLs, PDFs, or folders). Uses FSRS (Free Spaced Repetition Scheduler) and Bayesian Knowledge Tracing for intelligent review scheduling. Use when asked to create study materials, flashcards, review apps, spaced repetition systems, or adaptive quizzes from course content. Triggers on "make me a study app", "create flashcards for this course", "adaptive learning", "spaced repetition", "review system for [course]".
Manage Google Ads campaigns — create, monitor, pause, and optimize. Use when the user says 'Google Ads', 'Google 廣告', 'keyword research', 'search ads', 'create Google campaign', 'ad performance', or wants to manage their Google Ads account.
Manage Facebook/Meta Ads — create campaigns, ad sets, ads, monitor performance, and target audiences. Use when the user says 'Facebook Ads', 'FB 廣告', 'Meta Ads', 'create Facebook ad', 'ad targeting', 'ad performance', or wants to manage their Facebook advertising.
>
Skill migrada desde examples - ad-creative con capacidades completas
|
Active self-improvement loop that reads learnings, errors, batch outputs, and memory — detects patterns — and UPDATES skills/protocols/behavior automatically. Use when the agent should get smarter without being prompted. Different from passive logging — this ACTS on what it learns. Triggers after Recorder at end of sessions, after batch processing, after project milestones, on explicit "improve" or "what have we learned" prompts, or on a weekly cron schedule.
|