Skill UI

This tool provides a granular, per-message cost breakdown within a single conversation session. It serves as a deep-dive utility following an anomaly detection, allowing operators to pinpoint exactly which messages incurred the highest costs. It reveals critical cost components like output tokens, cache writes, and model escalations, helping users understand the true resource consumption beyond simple token counts.

Build a production-ready, AI-powered data collection agent for any public source (job boards, prices, news, etc.). This agent automatically scrapes data, enriches it using free LLMs (like Gemini Flash), and stores the results in databases such as Notion, Sheets, or Supabase. It runs on a free schedule using GitHub Actions and learns from user feedback, making it ideal for continuous data monitoring and tracking.

A comprehensive methodology for accelerating large-scale data ingestion, ETL processes, backfilling, and data warehouse loading. It guides users in identifying and optimizing bottlenecks—such as source extraction, network transfer, and loading speed—while strictly ensuring data correctness and maintaining clear accounting checkpoints throughout the pipeline.

This tool builds a comprehensive, graph-structured dossier on a seed entity. It performs parallel and recursive fan-out investigation across diverse sources, including the web, memory stores, knowledge graphs, and codebases. Ideal for mapping all connected entities and their provenance when a specific question is not defined.

This tool detects and classifies anomalies (such as spikes, drift, flatlines, or pattern breaks) in the telemetry data streamed from Cognitum Seed IoT devices. It utilizes Z-score analysis to quantify the deviation from normal operational metrics. Use this when investigating unexpected device behavior, validating stability before firmware advancements, or triaging large-scale fleet health alerts.

A read-only tool designed to compare prediction-market baskets or themes against the user's existing knowledge base, portfolio notes, research memos, or financial context. It performs gap analysis, flagging potential conflicts, stale assumptions, and areas of missing research. This skill provides structured insights into fit and exposure without offering any investment or trading advice.

This skill provides comprehensive, read-only market intelligence for prediction markets. It helps users research market events, compare different venues (like Polymarket, Kalshi), analyze liquidity, and understand underlying assets. It generates source-grounded briefings, emphasizing context and data quality, and strictly avoids providing investment or trading advice.

This tool performs deep pathfinding traversal within a knowledge graph, starting from a specified seed entity. It expands outwards through causal edges, calculates the cumulative relevance score for each path, and synthesizes the top results into a comprehensive report. Ideal for performing impact analysis, mapping complex dependency chains, or understanding abstract system relationships.

This framework manages the ingestion, organization, synchronization, and search of knowledge across multiple storage layers, including local files, persistent memory, vector stores, and Git repositories. It provides a structured approach for ensuring knowledge durability, deduplication, and maintaining a 'single source of truth' across complex development and operational contexts.

A comprehensive skill designed to fetch raw market data (Open, High, Low, Close, Volume) for a given symbol. It normalizes the time series data into relative OHLCV vectors, converts these vectors into 64-dimension embeddings, and stores them in a memory store. Crucially, it indexes the resulting vectors using the HNSW algorithm, enabling fast and precise nearest-neighbor search for pattern detection and similarity analysis in financial time series.

This powerful tool scans ingested OHLCV market data to detect and classify known candlestick patterns, such as Doji, Hammer, Engulfing, and Head & Shoulders. It analyzes single, multi-, and complex candle sequences, assigning a pattern name, type (reversal/continuation), direction, and a reliability score (0.0-1.0). Ideal for technical traders needing structured pattern identification for advanced market analysis.

This tool aggregates various system metrics—including counters, gauges, and histograms—from the observability namespace. It calculates baselines, flags significant deviations (anomalies) based on statistical standards, and provides a comprehensive system health report. Use it to get a detailed snapshot of system performance, monitor for degradation, and ensure overall system stability.