Results for “memory”

A workflow skill that lets Codex keep coding for 60 hours on medium and large projects: Claude Code CLI plans through an interactive requirements questionnaire, PRD, specs, and roadmap; Codex implements verified vertical slices, with resumable Long-Run Execution for lengthy work; Trellis maintains project memory. Use when the user mentions 大项目工作流, Codex 连续开发 60 小时, Codex 长任务, long-running or resumable Codex implementation, large/medium project, new SaaS, multi-module system, requirements questionnaire, Claude Code CLI planning, Codex implementation handoff, Trellis bootstrap, non-system-drive project, or vertical slice. Skip for: single-file edits, casual Q&A, copy polish, throwaway prototypes, or tasks the user explicitly wants done fast without process.

The English summary is being prepared. Showing the original description.

Expert in designing and building autonomous AI agents. Masters tool use, memory systems, planning strategies, and multi-agent orchestration.

The English summary is being prepared. Showing the original description.

Makes sessions durable across `/clear`, machine reboots, and IDE switches. SessionStart catchup pulls scratch + last MemPalace recall + planning files into context. Stop promotion writes a handoff note and runs memory-keeper. Manual `/mddesign:handoff` re-runs either at any time.

The English summary is being prepared. Showing the original description.

Turns a plan phase into a PhaseSpec v1 and dispatches it to a specialist subagent via the Agent tool. Subagent reads files itself, returns a structured result, the result is logged into progress.md. Markdown is the shared state; no in-memory shared store across agents.

The English summary is being prepared. Showing the original description.

Measure real frontend performance — Core Web Vitals (LCP, CLS, INP), page-load traces, CPU and network throttling, Lighthouse accessibility/SEO scores, and heap growth — using chrome-devtools-mcp against a live browser. Trigger when asked "is the page fast enough", "measure Core Web Vitals", "why is this route slow", "run Lighthouse", "check for a memory leak", before any performance-gated certification (Built for Shopify, Core Web Vitals thresholds, store listing review), and before any launch or traffic-multiplier event. Measurement only — this skill never drives functional journeys.

The English summary is being prepared. Showing the original description.

Use when claiming 'done', suggesting a commit, push, or PR — runs the evidence gate so completion claims come from fresh output in this message, not memory or earlier runs.

The English summary is being prepared. Showing the original description.

Before stating any specific fact about an LLM provider's product — models, pricing, limits, context windows, SDK/API — for OpenAI, Gemini, Claude & others, verify against official docs, not memory.

The English summary is being prepared. Showing the original description.

Use after an incident or outage is resolved — blame-free facilitation, root cause, corrective actions, memory write-back — even for a near-miss. Consumes the incident-commander skeleton.

The English summary is being prepared. Showing the original description.

Claude Code workflow tips from Boris Cherny (creator of Claude Code) and the Claude Code team. Use when: setting up Claude Code, optimizing workflows, running parallel sessions, configuring CLAUDE.md, using skills/commands, subagents, hooks, MCP integrations, or learning best practices. Covers: git worktrees, plan mode, verification, permissions, Slack MCP, BigQuery, prompting tips, plugins, custom agents, sandboxing, keybindings, status lines, output styles, customization, /simplify for code quality, /batch for parallel code migrations, /loop for scheduled tasks, code review agents, /btw for mid-task questions, /effort max reasoning, remote control sessions, voice mode, setup scripts, session naming, /color, PostCompact hook, auto mode, /schedule cloud jobs, iMessage plugin, auto-memory, and auto-dream, mobile app, session teleporting, Cowork Dispatch, Chrome extension, Desktop app, /branch forking, --bare SDK startup, --add-dir multi-repo, --agent custom agents, /voice input.

The English summary is being prepared. Showing the original description.

Archive merged feature specs into project memory with provenance, sweep discovery, and gated cleanup

The English summary is being prepared. Showing the original description.

Read the project's knowledge record before answering from memory or from the files. Use before answering about the project's state, a number, a date, a decision, a source, a deadline or what was agreed - even a casual 'what's the situation with X' or 'is that quote still current'; before changing a recorded value; and whenever the opener or a grounding line names entries. After finding relevant IDs, prefer kpop context to read them with their declared dependencies and checks. Uses pull for concise readings or when the checked reader is unavailable, affects for changed inputs, and search when no ID is known. A missing id is not absence. Not for showing the record's page (the hub skill). Requests come in any language.

The English summary is being prepared. Showing the original description.

[Debugging] Use when a workflow step or the user asks for performance bottleneck analysis and validated fixes: queries, memory, network, rendering and concurrency.

The English summary is being prepared. Showing the original description.

Build AI agents with Hugging Face's SmolAgents framework. Use when creating code-executing agents, tool-calling agents, multi-agent systems, agentic RAG, text-to-SQL pipelines, web browsing agents, or any multi-step AI workflows. Covers CodeAgent, ToolCallingAgent, custom tools, MCP integration, memory management, secure code execution (E2B, Docker, Blaxel), and model configuration (HF Inference, LiteLLM, Transformers, Ollama).

The English summary is being prepared. Showing the original description.

Sync reusable frontend decisions into persistent project documentation after UI work. Use when frontend changes introduce or modify component APIs, variants, design-system tokens, styling conventions, layout behavior, accessibility rules, interaction patterns, responsive behavior, or agent operating instructions; when the user asks to reflect decisions in components.md, design system docs, AGENTS.md, CLAUDE.md, GEMINI.md, or similar project memory files; or before finalizing frontend work where future consistency depends on decisions made during implementation.

The English summary is being prepared. Showing the original description.

Construa, depure e otimize aplicativos com Claude API / SDK Anthropic. Aplicativos construídos com esta skill devem incluir cache de prompt. Também lida com migração de código Claude API existente entre versões de modelo (4.5 → 4.6, 4.6 → 4.7, substituições de modelo descontinuado). TRIGGER quando: código importa `anthropic`/`@anthropic-ai/sdk`; usuário solicita Claude API, SDK Anthropic ou Managed Agents; usuário adiciona/modifica/afina um recurso Claude (cache, thinking, compaction, tool use, batch, files, citations, memory) ou modelo (Opus/Sonnet/Haiku) em um arquivo; dúvidas sobre prompt caching / cache hit rate em um projeto SDK Anthropic. SKIP: arquivo importa `openai`/SDK de outro provedor, nome de arquivo como `*-openai.py`/`*-generic.py`, código agnóstico de provedor, programação geral/ML.

The English summary is being prepared. Showing the original description.

Quantização de pesos consciente da ativação para compressão de LLM de 4-bit com aceleração 3x e perda mínima de acurácia. Use ao implantar modelos grandes (7B-70B) em memória GPU limitada, quando precisa de inferência mais rápida que GPTQ com melhor preservação de acurácia, ou para modelos com instruções ajustadas e multimodais. Vencedor do Prêmio Melhor Artigo MLSys 2024.

The English summary is being prepared. Showing the original description.

Orientação especializada para fine-tuning rápido com Unsloth - treinamento 2-5x mais rápido, 50-80% menos memória, otimização LoRA/QLoRA

The English summary is being prepared. Showing the original description.

Quantiza LLMs para 8-bit ou 4-bit com redução de memória de 50-75% e perda mínima de acurácia. Use quando a memória GPU é limitada, precisa ajustar modelos maiores ou quer inferência mais rápida. Suporta formatos INT8, NF4, FP4, treinamento QLoRA e otimizadores 8-bit. Funciona com HuggingFace Transformers.

The English summary is being prepared. Showing the original description.

Ajuste fino com eficiência de parâmetros para LLMs usando LoRA, QLoRA e 25+ métodos. Use ao fazer fine-tuning de modelos grandes (7B-70B) com memória GPU limitada, quando precisa treinar <1% dos parâmetros com perda mínima de precisão, ou para serviços multi-adapter. Biblioteca oficial do HuggingFace integrada ao ecossistema transformers.

The English summary is being prepared. Showing the original description.

Otimiza atenção em transformers com Flash Attention para ganho de 2-4x em velocidade e redução de 10-20x em memória. Use ao treinar/executar transformers com sequências longas (>512 tokens), ao encontrar problemas de memória GPU com atenção, ou quando precisa de inferência mais rápida. Suporta SDPA nativo do PyTorch, biblioteca flash-attn, H100 FP8 e sliding window attention.

The English summary is being prepared. Showing the original description.

Quantização Half-Quadratic para LLMs sem dados de calibração. Use ao quantizar modelos com precisão 4/3/2-bit sem necessidade de conjuntos de calibração, para workflows de quantização rápida, ou ao fazer deploy com vLLM ou HuggingFace Transformers.

The English summary is being prepared. Showing the original description.

Quantização pós-treinamento em 4-bits para LLMs com perda mínima de precisão. Use para implantar modelos grandes (70B, 405B) em GPUs de consumo, quando você precisa de redução de memória 4× com <2% de degradação de perplexidade, ou para inferência mais rápida (aceleração de 3-4×) vs FP16. Integra com transformers e PEFT para fine-tuning QLoRA.

The English summary is being prepared. Showing the original description.

Sistemas de memória persistente para conversas com LLM, incluindo memória de curto prazo, longo prazo e baseada em entidades. Use quando: memória de conversa, lembrar, persistência de memória, memória de longo prazo, histórico de chat.

The English summary is being prepared. Showing the original description.

Memória é a base de agentes inteligentes. Sem ela, cada interação começa do zero. Esta skill cobre a arquitetura de memória de agentes: curto prazo (janela de contexto), longo prazo (vector stores), e as arquiteturas cognitivas que as organizam. Insight-chave: Memória não é apenas armazenamento - é recuperação. Um milhão de fatos armazenados não significam nada se você não conseguir encontrar o certo. Estratégias de chunking, embedding e recuperação determinam se seu agente se lembra ou esquece. O campo é fragm

The English summary is being prepared. Showing the original description.