Skip to main content

search

Search for papers, projects, ecosystem players, or solutions on a topic

الانتقال إلى التثبيت

معلومات المصدر

المستودع
redhat-et/physical-ai-platform-intel
آخر نشاط في المصدر
١٦ يونيو ٢٠٢٦ في ٠٦:٥٣
لغة SKILL.md المكتشفة
الإنجليزية
النجوم
١
التفرعات
٢

خيارات التثبيت

يُحدَّد Prompt الذي يراجع المصدر أولًا بشكل افتراضي. يمكنك التبديل إلى أمر مباشر أو تنزيل نسخة محلية.

مراجعة ملفات المصدر

اقرأ SKILL.md وأي ملفات مرافقة يعرضها SkillsMP قبل أن تقرر التثبيت.

عرض SKILL.md

SKILL.md
تعليمات المصدر · معاينة للقراءة فقط
name
search
description
Search for papers, projects, ecosystem players, or solutions on a topic
user-invocable
true
argument-hint
<topic>
# Search for Content Search for papers, projects, ecosystem players, or solutions on a specific topic. ## Usage ```text /search <topic> ``` ## Examples ```text /search JEPA /search robot foundation models /search digital twin open source /search Isaac ROS competitors /search Physical AI startups funding 2026 ``` ## Instructions When this command is invoked: 1. **Parse the topic** to understand what to search for 2. **Determine search strategy**: - For papers: Use WebSearch with arXiv and Google Scholar - For projects: Search GitHub via WebSearch - For ecosystem players/solutions: Search vendor blogs, TechCrunch, The Robot Report, Crunchbase - For general content: Multi-source search 3. **Execute searches**: - **arXiv + TechRxiv**: Search with topic keywords + filter last 3-6 months - **Google Scholar**: For applied/engineering papers - **GitHub**: Search repos matching topic with stars>10 - **Research blogs**: Research lab blogs (Meta AI, Google DeepMind, etc.) - **Industry sources**: TechCrunch, The Robot Report, IEEE Spectrum for startups and industry news - **Vendor pages**: Product documentation and developer portals - Use search queries from `research/tools/preferred-sources.md` 4. **Rank and filter results**: - Relevance to Physical AI platform intelligence - Recency (prefer last 6 months for papers) - Quality indicators (citations, stars, author reputation, deployment evidence) - Avoid duplicates already in research documents 5. **Present findings** to user: - Show top 5-10 results - Include: title, source, date, 1-sentence description - Format as numbered list for easy selection - Group by category: Papers, Projects, Ecosystem Players, Solutions 6. **Ask which to add**: "Which of these would you like me to add? (e.g., '1, 3, 5')" 7. **Add selected items** using the `/add` workflow for each ## Search Sources Priority 1. **arXiv + TechRxiv** - Research papers (foundational and applied) 2. **Google Scholar** - Applied/engineering papers, conference proceedings 3. **GitHub** - Open-source projects and implementations 4. **Industry news** - The Robot Report, TechCrunch, IEEE Spectrum for startups and product launches 5. **Company blogs** - Research labs and vendor product announcements 6. **Crunchbase** (via WebSearch) - Startup funding and company profiles 7. **Researcher pages** - Latest work from key people Refer to `research/tools/preferred-sources.md` for specific search queries and sources. ## Output Format ```text Found 10 results for "robot foundation models": Papers: 1. [2026-05] "OpenPI: Open-Source Policy for Robot Manipulation" - Physical Intelligence arXiv:2605.xxxxx - Open VLA model achieving 85% success on 24 tasks 2. [2026-04] "GR00T N1.7: Generalist Robot Policy" - NVIDIA arXiv:2604.xxxxx - Multi-embodiment foundation model Projects: 3. [Active] Physical-Intelligence/openpi - Official OpenPI implementation (2.1k stars) Python/JAX - Open-source VLA with gRPC serving Ecosystem: 4. [Startup] Skild AI - $300M Series A, building general-purpose robot FM Founded 2023, CMU spinout, claims 4x data efficiency Solutions: 5. [Product] NVIDIA Isaac Manipulator - Manipulation pipeline on Isaac ROS GPU-accelerated grasping, integrates with Omniverse Which would you like me to add? (e.g., '1, 3, 5') ``` ## Error Handling - If no results found, suggest related search terms - If too many results (>20), ask user to narrow the search - If search fails, try alternative search methods
عرض على GitHub