decay-scoring
Sistema de scoring temporal para nodes de conhecimento do corpus RAG. Calcula scores de obsolescencia baseado em idade, validacao e acesso. Gera sugestoes de curadoria para conteudo desatualizado.
Sistema de scoring temporal para nodes de conhecimento do corpus RAG. Calcula scores de obsolescencia baseado em idade, validacao e acesso. Gera sugestoes de curadoria para conteudo desatualizado.
Create, clean, and optimize datasets for LLM fine-tuning. Covers formats (Alpaca, ShareGPT, ChatML), synthetic data generation, quality assessment, and augmentation. Use when preparing data for training.
Specialized research expert for parallel information gathering. Use for focused research tasks with clear objectives and structured output requirements.
Intelligently choose between semantic and text search based on query intent. Automatically selects the best search mode (semantic for concepts, text for exact terms, symbols for definitions) and provides relevant results. Use when user wants to find code.
Document citations and RAG (Retrieval-Augmented Generation) patterns for Claude. Activate for source attribution, document grounding, citation extraction, and contextual retrieval.
Implement comprehensive mathematical theorem proving capabilities with SFT+GRPO training, MCP/A2A agent integration, and imatrix quantization protection to surpass Boreas-phi3.5-instinct-jp in formal proof generation and scientific discovery. Use when building mathematical reasoning systems, formal verification tools, or AI-assisted theorem proving environments.
Gemini File Search (RAG) - semantic search over your documents
Provide accurate, evidence-based answers by referencing sources from documents, textbooks, or research papers. Use when user asks for factual or source-backed responses.
Look up current research information using Perplexity Sonar Pro Search or Sonar Reasoning Pro models through OpenRouter. Automatically selects the best model based on query complexity. Search academic papers, recent studies, technical documentation, and general research information with citations.
Read and summarize constitution principles for the AI-Native Robotics Textbook project. Use when checking project rules, validating compliance, or understanding constraints.
The "Recall Engine". Fetches history, decisions, and tasks related to a person or topic.
Automatically fetch and summarize the latest AI news, research, and industry developments. Use when users request: (1) Daily AI news updates, (2) Latest AI technology developments, (3) Recent AI research papers or breakthroughs, (4) AI industry trends and market news, (5) Specific AI company announcements, or (6) Automated daily AI briefings. Supports customizable search queries, multi-source aggregation, and formatted output in various styles.
Recall and save insights using sage_recall_knowledge and sage_save_knowledge MCP tools. INVOKE WHEN: user asks "what do we know about", "recall", "remember this"; starting research on a topic that may have prior knowledge; user wants to save an insight for future use. DO NOT INVOKE: for general questions without stored context.
Generates a comprehensive learning graph from a course description, including 200 concepts with dependencies, taxonomy categorization, and quality validation reports. Use this when the user wants to create a structured knowledge graph for educational content.
Build Retrieval-Augmented Generation (RAG) Q&A systems with Claude or OpenAI. Use for creating AI assistants that answer questions from document collections, technical libraries, or knowledge bases.
Production-grade text search algorithms for finding and matching text in large documents with millisecond performance. Includes Boyer-Moore search, n-gram similarity, fuzzy matching, and intelligent indexing. Use when building search features for large documents, finding quotes with imperfect matches, implementing fuzzy search, or need character-level precision.
This skill should be used when working with genomic interval data (BED files) for machine learning tasks. Use for training region embeddings (Region2Vec, BEDspace), single-cell ATAC-seq analysis (scEmbed), building consensus peaks (universes), or any ML-based analysis of genomic regions. Applies to BED file collections, scATAC-seq data, chromatin accessibility datasets, and region-based genomic feature learning.
This skill automatically generates a comprehensive glossary of terms from a learning graph's concept list, ensuring each definition follows ISO 11179 metadata registry standards (precise, concise, distinct, non-circular, and free of business rules). Use this skill when creating a glossary for an intelligent textbook after the learning graph concept list has been finalized.
Web search and content extraction toolkit. Use for searching documentation, facts, current information, or extracting readable content from URLs. Supports multiple providers (ddgs keyless, brave_api with key), caching, and safe defaults. Prefer this over browser-tools when no interaction is needed.
Multi-provider research lookup supporting Gemini Deep Research (60-min comprehensive analysis) and Perplexity Sonar (fast web-grounded research). Intelligently routes between providers based on research mode and query complexity. Supports balanced mode for optimal quality/time tradeoff.