A reasoning-only skill that converts user-provided advertising performance data into decisions about what to stop, scale and test. It compares metrics against benchmarks, identifies wasted spend and winners, checks the statistical significance of A/B tests, analyzes the conversion funnel and compares channel efficiency. It then proposes budget reallocation with projected scenarios and an action plan.
Guides an agent to add educational comments to a user-selected code file. Comments explain syntax choices, patterns and design decisions, adapting their depth to beginner, intermediate or advanced knowledge. It preserves encoding, line endings, indentation and syntactic validity. The target is 125% of the original file length, capped at 400 new lines. If no file is specified, the agent asks for one and lists matching files by number.
Reviews Pull Requests in awesome-copilot using editorial judgments that automated schema validation cannot settle: repository fit, actual knowledge value added to the model (high, partial or none), differentiation from existing resources, testing evidence and commercial promotion. Changes to review policy files are treated as security-sensitive.
Helps an agent write internal communications in defined formats, including 3P updates (Progress, Plans, Problems), company newsletters, FAQs, status reports, leadership and project updates, and incident reports. It identifies the communication type, loads the appropriate guidance from examples/, and follows its formatting, tone and information-gathering instructions. Slack, Google Drive, email and calendar can be used when available.
Guidance for designing new web interfaces or reshaping existing ones with a deliberate, distinctive visual identity. It covers an aesthetic direction grounded in the product and audience, design tokens for colors, typography and layout, and reviewing the plan against the brief to avoid common AI-generated design patterns. It also addresses restrained motion, accessibility, mobile responsiveness and clear interface copy written from the user's perspective.
Guides an assistant to append two or three short, specific follow-up questions after substantive answers a user may act on, including advice, estimates, plans, drafts, data analysis and factual claims. Questions refer to concrete parts of the answer and help check facts, reasoning, assumptions and missing context. The nudge appears at most once per conversation and is skipped for simple requests or when the user explicitly asks to omit it.
Applies Anthropic's official visual identity to outputs such as presentations and documents. The base colors are Dark #141413, Light #faf9f5, Mid Gray #b0aea5 and Light Gray #e8e6dc. Accent colors are Orange #d97757, Blue #6a9bcc and Green #788c5d. Headings use Poppins and body text uses Lora, with Arial and Georgia as fallbacks. Text color adapts to the background, while non-text shapes cycle through the accent colors.
Start with the result you want, then choose a skill for your agent.
02
Understand the requirements
Read the guide, limitations and permissions before adding a skill.
03
Download and try
Download the reviewed version and follow your agent’s guide. Start with a small task in a test project.
Files are reviewed before publication, but no scan guarantees absolute safety. Limit access to secrets and permissions, and review commands before running them.
مساعد دروب أيديا
مساعدك الذكي · هنا لمساعدتك
لنبدأ التعارف
أهلاً بك في دروب أيديا
عرّفنا بنفسك لنساعدك بشكل أفضل، وليتمكن فريقنا من متابعتك عند الحاجة.
بياناتك تُستخدم للتواصل فقط ولا تُشارَك.
من الفكرة إلى الخطوة التالية
كيف تحب نتواصل؟
اسأل عن خدماتنا أو احكِ لنا عن مشروعك. اختر الطريقة الأنسب لك، ولنبدأ.
المحادثة الصوتية غير مدعومة في هذا المتصفح، استخدم المحادثة الكتابية.
مساعد بالذكاء الاصطناعي، وفريقنا حاضر عند الحاجة.
جارٍ التحضير…
فريقنا جاهز لمساعدتك
اترك بياناتك وسيتواصل معك أحد أعضاء فريق دروب أيديا في أقرب وقت.