В каталоге 182 820 навыков в 14 категориях.
Use when a user has a rough prompt that needs a staged prompt-improvement workflow for `$ralph-wiggum-codex`, with saved...
Generate Ralph-compatible prompts for multiple related tasks. Creates phased prompts with sequential milestones, cumulat...
Generate Ralph-compatible prompts for entire projects from scratch. Creates comprehensive prompts with architecture phas...
Generate Ralph-compatible prompts for research, analysis, and planning tasks. Creates prompts with systematic research p...
Generate Ralph-compatible prompts for single implementation tasks. Creates prompts with clear completion criteria, autom...
Resume a PAUSED Ralph session — retry failed task, skip, or abort
Resume from saved session state after interruption or compaction. Recovers context from checkpoints and agent-memory.
Run RALPH autonomous development loop to implement features from the PRD.
Run long-running autonomous AI agents using PRD-based task scoping.
Self-improving review loop for Ralph Wiggum skills. Reviews skills against best practices, implements improvements, and...
Spawn parallel agents in git worktrees for automated feature implementation. Creates isolated development environments f...
Use only when the user explicitly invokes `$ralph-specum`, requests Ralph Specum in Codex, asks Ralph Specum to handle a...
This skill should be used only when the user explicitly asks to use `$ralph-specum-cancel`, or explicitly asks Ralph Spe...
This skill should be used only when the user explicitly asks to use `$ralph-specum-help`, or explicitly asks Ralph Specu...
This skill should be used only when the user explicitly asks to use `$ralph-specum-switch`, or explicitly asks Ralph Spe...
Monitor all active Ralph Wiggum agents and their progress. Shows iteration counts, gate status, and blocked tasks.
The Ralph Wiggum technique for minimal, declarative prompts that enable loop-based autonomous execution.
Run ralph-tui task loops for automated PRD execution. Covers headless mode, PRD preparation, agent selection (opencode/c...
AI agent loop orchestrator for autonomous task execution - coding, content, any multi-step work
Long-running iterative development loops with pacing control and verifiable progress. Use when tasks require multiple it...
Ralph Wiggum iterative development loop methodology for persistent AI agent work. Implements continuous iteration loops...
Autonomous AI coding loop for unattended development - lets Claude work continuously on tasks while you sleep, with mand...
Executa loops autonomos de desenvolvimento via tecnica Ralph Wiggum. Usar quando: (1) Iniciar loop Ralph na sessao atual...
Surgical Debugger & Code Optimizer. Autonomous root-cause investigation, persistence-loop fixing, and high-fidelity code...
Use when a coding task needs an objective-first long-running loop with fresh-context work and review phases, explicit ac...
Multi-step autonomous plan execution with bounded iteration (max 10), state persistence, dependency checking, error retr...
Ralph Wiggum: Project Manager — Continuous ticket consumption, code review, and work generation loop. Picks up the next...
Autonomous TDD development loop with parallel agent swarm, category evolution, and convergence detection. Use when runni...
Autonomous task completion loop with PRD tracking and LLM memory
Next-generation autonomous development orchestrator with cognitive feedback loops. Executes complex multi-step features...
Autonomous AI coding orchestration using the Ralph Wiggum technique and Ralphy CLI. This skill should be used when runni...
AgentDB integration specialist for RAN ML systems with vector storage, pattern recognition, and distributed training coo...
Causal inference and discovery for RAN optimization with Graphical Posterior Causal Models (GPCM), intervention effect p...
DSPy-based mobility optimization with temporal patterns, handover management, and 15% improvement target. Uses program s...
Comprehensive RAN optimization with swarm coordination, cognitive consciousness, and 15-minute closed-loop autonomous cy...
Reinforcement learning engineering for RAN systems with policy gradients, experience replay, and AgentDB integration. Im...
Builds a fast, working, credible-enough prototype to prove an AI opportunity using AI-assisted coding ("vibe coding") —...
Interest rate cycle analysis, Fed policy analysis, yield curve dynamics, duration management, central bank rate trajecto...
Ray distributed computing framework. Use for scaling ML.
Distributed Python compute with Ray — @ray.remote tasks/actors for cluster-scale parallelism, Ray Data for large-batch p...
Framework for scaling Python applications from a laptop to a cluster. Includes Ray Core for distributed computing, Ray S...
Comprehensive guide to using Ray for scalable distributed computing, including Ray Core, Data, Train, Tune, Serve, and R...
Distributed computing skill using Ray for parallel training, hyperparameter search, and resource management.
Distributed training orchestration across clusters. Scales PyTorch/TensorFlow/HuggingFace from laptop to 1000s of nodes....
Cheminformatics toolkit for fine-grained molecular control. SMILES/SDF parsing, descriptors (MW, LogP, TPSA), fingerprin...
a very basic but solid chatbot agent in your terminal
AI 模型安全评估与取证（行为层）：extraction assessment（API 黑盒提取评估）、 fingerprint verification（行为指纹与归属验证）、privacy leakage evaluation（成员推断...
AI 模型分析第一入口（分流器）：识别输入形态——模型文件走文件层逆向（re-ai-model）、 仅有 API 走行为层评估（re-ai-attack）、恶意行为走恶意分析。 触发词：AI模型分析、AI模型安全、模型文件、模型泄露、模型逆...
Python 打包/混淆样本分析：PyInstaller/PyArmor/Nuitka/Cython 解包、pyc 版本识别与反编译。 触发词：Python打包、PyInstaller、PyArmor、pyc、python exe、Pyth...
Física de muelles en React con @react-spring/web (useSpring, useTrail, useTransition, useScroll, useInView) y Popmotion...
Source code migration engine. Rewrites every deprecated React pattern to React 19 APIs - forwardRef, defaultProps, React...
Read and analyze Inspect AI evaluation log files using the Python API. Extract samples, messages, events, and metrics fr...
Chat with Readwise highlights using GPT-5.1 RAG. Use when the user wants to ask questions about their reading library, g...
国内房产投资(房产投资人视角 — 城市与地段选择 / 政策与调控解读 / 一手房与二手房 / 房产金融与杠杆 / 估值与尽职调查 / 交易流程与税费 / 持有运营与出租 / 退出与流动性 / 风险识别 / 房产在个人资产配置中的位置 / 投...
REANN 势能面训练技能，基于递归原子神经网络拟合势能面。适用于：机器学习力场训练、势能面拟合、分子动力学前置模型训练与包含 REANN 标准训练包的数据任务。
Reasoning and planning framework. Defines task analysis, risk assessment, hypothesis forming. Applied automatically for...
Multi-provider reasoning/thinking configuration. Discriminated unions for OpenAI effort, Anthropic budget, Google level/...
Meta-skill that wraps reasoning skills to capture notes in real-time, protecting against context compression. Creates ch...
Use this skill for rigorous theoretical derivation with supercollider mode (G1-G7 simultaneous), diffusion reasoning, an...
Implement ReasoningBank adaptive learning with AgentDB's 150x faster vector database. Includes trajectory tracking, verd...
Implement adaptive learning with ReasoningBank for pattern recognition, strategy optimization, and continuous improvemen...
Restore context and momentum via Pratyabhijñā (recognition)
Reconstruct and narrate the current development context from contextual commits. Run at session start, when resuming wor...
Query memory system for relevant learnings from past sessions
Search Origin's local memory by query. Targeted lookup, not orientation. Invoked as `/recall <query>`. Use when the user...
Deep search across all past Claude Code sessions for decisions, solutions, and discussions
Query the memory system for relevant learnings from past sessions
Quick single-project context check. Reads next-session.md directly (no subagent) and outputs a 6-field briefing. For mul...
Query the memory system for relevant learnings from past sessions using semantic search.
Search persistent memory, git history, and dev_env SSOT before work. Use at session start, when taking a new task, or wh...
Search past reasoning for relevant decisions and approaches
Recall the reasoning behind a past change by locating the Claude Code transcript that produced it. Use when the user ask...
Use when a task involves continuing a project, restoring project context, maintaining file-based project memory, updatin...
FREE — God-tier long-context memory for AI agents. Injects 500K-1M clean tokens, auto-summarizes with tone/intent preser...
Get session recap including preferences, recent context, and active goals. Use at the start of a session or when context...
回顾最近 N 天的 Claude Code 使用记录——扫描原始会话数据，按主题分组汇总"我都做了什么"，并从个人操作系统视角输出模式、风险与增删建议。当用户说 /recap、"看看我这几天做了什么"、"回顾一下我最近的会话"、"这两天我用...
Use Recent Data to find and apply recently accessed topics and resources.
Guide for using Recent Data (topics, resources, pages). Use when working with recently accessed items, implementing rece...
Find and quote statutory and regulatory law with the LibreJustice MCP tools — French codes at any date, EU law, treaties...
Find and quote legal texts and administrative guidance with the LibreJustice MCP tools, covering French codes at any dat...
Refactor CLAUDE.md files to follow progressive disclosure principles. Use when CLAUDE.md is too long or disorganized.
Reads ~/worklog journal pages to recollect todos, insights, and open problems. Use for "what was I doing", catch-up, pos...
Build recommendation systems with collaborative filtering, matrix factorization, hybrid approaches. Use for product reco...
Deploy production recommendation systems with feature stores, caching, A/B testing. Use for personalization APIs, low la...
Review a worker session summary and reconcile beads tasks with implementation reality. Use after a worker completes /fin...
使用 Playwright 进行浏览器自动化录屏。当需要录制 Web 应用操作演示、生成带时间线的屏幕录像时使用。
Polish screen recordings with Recordly (open-source Screen Studio alternative) — auto-zooms, cursor animations, webcam o...
Recover session context.
[Utilities] Restore workflow context from checkpoint after session loss
Get up to speed on the current branch's intent by analyzing commits, changes, and optionally a Linear ticket. Use when s...
Recover from missing workspace context.
Designs composable recommendation, ranking, and feed pipelines using the six-stage Source→Hydrator→Filter→Scorer→Selecto...
Optimizes AI agent performance by pruning redundant context, managing token usage, and enforcing ultra-concise, direct-t...
Self-improvement engine. Implements generate-critique-iterate loops for enhanced reasoning. Use when working through com...
Orchestrate multi-agent collaborative document synthesis through 6 phases - Divergence, Synthesis, Commentary, Consolida...
Designs self-referential and recursive systems that examine, modify, or generate themselves, including metacognitive arc...
Use this to adversarially test an LLM/agent app before attackers do - prompt injection, jailbreaks, data exfiltration, t...
通过 AI 驱动的语义搜索，快速从 Reddit 的海量真实讨论中提取用户痛点、产品对比、趋势信号和市场空白，为产品创意验证、竞品分析和需求挖掘提供直接洞察。
红狐订阅中心 — 统一管理全部可订阅内容源：账号追踪、AI 日报、行业日报、热榜推送。用户说「我要订阅」时展示分组菜单并引导完成订阅；「我的订阅」汇总当前所有定时推送任务；支持一键取消订阅。当用户想要订阅日报、追踪账号更新、管理推送任务、查...
Redis LangCache guidance for semantic caching of LLM responses on Redis Cloud — calling search/set via the SDK or REST A...