Memory consolidation skill that replicates Anthropic's Auto Dream feature. Runs a 4-phase reflective pass over memory files: Orient → Gather → Merge → Prune. Use when: (1) Context window feels cluttered with stale info, (2) After long coding sessions, (3) Manually triggered with /dream, (4) Automatically after daily-reflection. Keeps memories tight, removes contradictions, converts relative dates to absolute.
Lena lernt aus jeder Konversation und verbessert sich automatisch
Use when calculating gear ratios, converting RPM between shafts, computing torque output, analyzing drivetrain configurations, or selecting motors for mechanical systems.
Complete cybersecurity assessment, threat modeling, and hardening system. Use when conducting security audits, threat modeling, penetration testing, incident response, or building security programs from scratch. Works with any stack — zero external dependencies.
Extract actionable insights from books using Four-Layer Methodology: (1) Skeleton - conceptual frameworks and mental models, (2) Flesh - 2-3 detailed case studies including original examples, cross-industry analogies, and real-world applications, (3) Essence - cross-industry migration matrices with specific industry adaptations and 3-5 step executable SOPs, (4) Residue - critical analysis of boundaries, limitations, and failure conditions. Dual processing modes: Quick (5 core points, 10-15 min) for rapid assessment and Deep (10-20 comprehensive points, 30-45 min) for systematic learning. Includes Feynman validation testing with scenario-based problems and scoring rubrics. Generates structured reports in Markdown/PDF/Word formats. Use when user requests systematic knowledge extraction, concept distillation, or implementation guidance from methodology/business/psychology/self-help books with emphasis on practical application and cross-domain transfer.
AI-powered fashion model pose transfer tool. Generate pose variations of a model/product image using reference pose images while keeping clothing and background consistent.
自我进化能力(Dream Module)，空闲时自动分析失败案例，学习新模式，更新Pattern库
Meta-cognitive self-learning system - Automated skill evolution based on predictive coding and value-driven mechanisms.
Safe local Mac optimization toolkit for OpenClaw agents on Apple Silicon. 1-trillion agent swarm simulation, local CoreML resource forecasting, safe cleanup & backups. 100% user-level, no internet, no persistence. Discoverable via ClawHub search.
Alias of air-train-ev. Unified travel + mobility skill: (1) flight pricing with Amadeus (flight offers), (2) public transport/train journey planning with Navitia (journeys, departures), and (3) find nearby EV charge points using Open Charge Map. Use when Alessandro asks for flight prices, train itineraries/schedules, or EV charging stations.
Use when evaluating, testing, and optimizing an agent architecture or multi-agent system. Best for reviewing planning, routing, memory, tool use, reliability, observability, cost, and system-level failure modes.
Complete machine learning pipeline for trading: feature engineering, AutoML, deep learning, and financial RL. Use for automated parameter sweeps, feature creation, model training, and anti-leakage validation.
NautilusTrader algorithmic trading platform for strategy development and live trading. Use when building trading strategies, backtesting, or deploying to Hyperliquid.
Scar memory, reflex arc, and decision traces for AI agents. Learn from failures permanently. Block repeated mistakes instantly — no LLM calls needed. Three-layer memory: scars (immutable failures) + narrative (overwritable) + decision traces (judgment paths → LoRA training data).
Train high-performance medical LLMs on consumer GPUs using parameter-efficient fine-tuning
3-tier Perplexity AI search routing with auto model selection
Live LLM model intelligence and pricing from OpenRouter
LLM-as-a-Judge evaluator via Langfuse. Scores traces on relevance, accuracy, hallucination, and helpfulness using GPT-5-nano as judge. Supports single trace scoring, batch backfill, and test mode. Integrates with Langfuse dashboard for observability. Triggers: evaluate trace, score quality, check accuracy, backfill scores, test evaluator, LLM judge.
LLM-as-a-Judge evaluation system using Langfuse. Score AI outputs on relevance, accuracy, hallucination, and helpfulness. Backfill scoring on historical traces. Uses GPT-5-nano for cost-efficient judging. Use when evaluating AI quality, building evals, or monitoring output accuracy.
Multi-model consensus system — send a query to 3+ different LLMs via OpenRouter simultaneously, then a judge model evaluates all responses and produces a winner, reasoning, and synthesized best answer. Like having a board of AI advisors. Use for important decisions, code review, research verification.
