Claude Skills by majiayu000

Optimize context windows and token usage for LLM efficiency.

Master context engineering for AI agents - token optimization, degradation patterns, compression, memory systems, multi-agent coordination, evaluation. Use when designing agents, debugging context failures, or building LLM pipelines.

Context optimization extends the effective capacity of limited context

Writes meaningful ai_context values in Crowdin JSONL files for strings that need translator context. Use when extracting context, filling ai_context fields, adding context to ambiguous strings, or working with Crowdin JSONL context files.

Phase 0.1 — Automatic codebase scan before PRD generation. Extracts key terms from the user prompt, searches the codebase for related files, identifies existing patterns, and generates context.md in the state directory.

Forces Claude to extract exact quotes from files before making claims about them. Prevents hallucinations by grounding every statement in actual text from the codebase. Use when working with large files or unfamiliar code.

Creates context handoff documents for session continuity. Use when ending sessions, switching tasks, or handing off work.

Creates context handoff documents for session continuity. Use when ending sessions, switching tasks, or handing off work.

Creates context handoff documents for session continuity. Use when ending sessions, switching tasks, or handing off work.

Design memory hierarchy with progressive loading for optimal context management. Use when organizing CLAUDE.md imports, implementing just-in-time context loading, or designing priming hierarchies for agents.

Use Context Hub (chub) for curated, versioned API/SDK docs and agent skills. Keep OpenAI tasks on official docs first.

Place trusted contextual information into prompts or agent state using explicit boundaries, provenance, and templates. Use when relevant context has already been selected and must be inserted safely; use context-retrieval to find it or context-optimization to choose and order it.

ALWAYS activate this skill at the start of every task. Load project context from .claude/ directory including coding rules, design decisions, and documentation before executing any task.

Load project context from .claude/ directory at the start of every task. This ensures Gemini CLI has the same coding rules, design decisions, and library constraints as Claude Code.

Use when working with context management context restore

Use when working with context management context save

Manage project context effectively. Use when onboarding, context switching, or maintaining project knowledge. Covers context capture and sharing.

Proactive context window management via token monitoring, intelligent extraction, and selective rehydration. Features predictive budget monitoring, context health indicators, and priority-based retention. Use when approaching token limits or needing to preserve essential context. Complements /transcripts and PreCompact hook with proactive optimization.

Proactive context window management via token monitoring, intelligent extraction, and selective rehydration. Features predictive budget monitoring, context health indicators, and priority-based retention. Use when approaching token limits or needing to preserve essential context. Complements /transcripts and PreCompact hook with proactive optimization.

Strategies for managing AI agent context windows including optimization, summarization, retrieval-augmented generation, progressive disclosure, and pruning. Use when the user is hitting context limits, building long-running agents, implementing RAG, optimizing token usage, or designing systems that need to manage large amounts of information within limited context windows.

Elite AI context engineering specialist mastering dynamic context management, vector databases, knowledge graphs, and intelligent memory systems.

Audit and manage the full project context landscape: CLAUDE.md memory hierarchy, project documentation, markdown footprint, and content overlap. Detects project type, scores quality, flags stale docs, and reports total context cost. Trigger with 'audit context', 'audit memory', 'update CLAUDE.md', 'restructure memory', 'session capture', 'check project docs', 'markdown footprint', or 'what docs does this project need'.

Audit and manage the full project context landscape: CLAUDE.md memory hierarchy, project documentation, markdown footprint, and content overlap. Detects project type, scores quality, flags stale docs, and reports total context cost. Trigger with 'audit context', 'audit memory', 'update CLAUDE.md', 'restructure memory', 'session capture', 'check project docs', 'markdown footprint', or 'what docs does this project need'.

Map relationships between bounded contexts using DDD context mapping patterns. Use when defining upstream/downstream relationships, integration strategies (ACL, OHS, PL), or generating Context Mapper DSL output. Follows event storming for bounded context discovery.

Advanced token optimization and context management strategies.

Entry point to the Context Mate toolkit - skills, agents, and commands that work with Claude Code's natural flow. Project lifecycle (/explore-idea → /plan-project → /wrap-session), session handoff across context windows, developer agents for specialized tasks, and quality auditing. Use when: starting new projects, understanding the toolkit, needing workflow guidance. "It's all about the context, maaate!"

Context window conservation rules. Invoke when approaching context limits or before large tasks.

Optimize context window via MECW principles, memory tiering, and subagent coordination.

Apply compaction, masking, and caching strategies to optimize context windows. Use when reducing token usage or improving LLM context efficiency.

This skill should be used when the user asks to "optimize context", "reduce token costs", "improve context efficiency", "implement KV-cache optimization", "partition context", or mentions context limits, observation masking, context budgeting, or extending effective context capacity.

This skill should be used when the user asks to "optimize context", "reduce token costs", "improve context efficiency", "implement KV-cache optimization", "partition context", or mentions context limits, observation masking, context budgeting, or extending effective context capacity.

'> [IMPORTANT] Use TaskCreate to break ALL work into small tasks BEFORE

Apply optimization techniques to extend effective context capacity. Use when context limits constrain agent performance, when optimizing for cost or latency, or when implementing long-running agent systems.

Strategies for managing Claude Code's context window efficiently: token budgeting, progressive disclosure, reference files, and using subagents to protect context. Use when conversations are getting long, the user mentions context limits, or tasks involve reading many files.

Context Optimization - Orchestrate RPI workflow TODO generation with interactive scoping

Use this skill when context is growing quickly or cache consistency drops.

Second-pass context optimization that analyzes user prompts and removes irrelevant specs, agents, and skills from loaded context. Achieves 80%+ token reduction through smart cleanup. Activates for optimize context, reduce tokens, clean context, smart context, precision loading.

上下文优化专家。专注于长对话中的上下文管理、token 效率和性能优化。解决 lost-in-middle、context poisoning 等问题，提升 AI 代理在复杂任务中的表现。

Proactively preserve critical information before automatic context pruning destroys it.

Runtime context passed to compiled closures — the world as seen from inside

Runtime context passed to compiled closures — the world as seen from inside

Context management tools for Claude Code - provides intelligent codebase mapping with Python, Rust, and C++ parsing, duplicate detection, and MCP-powered symbol queries. Use this skill when working with large codebases that need automated indexing and context management.

Master Claude Code's context primitives (Skill, Task, Command) and orchestration patterns. Use when designing agent workflows, understanding why some patterns work while others fail (e.g., Task→Task is forbidden but Skill→Skill is allowed), or implementing Chain of Experts architectures. Not for writing prompts or content generation.

Strategies for managing LLM context windows including summarization, trimming, routing, and avoiding context rot Use when: context window, token limit, context management, context engineering, long context.

Strategies for managing LLM context windows including

コンテキストウィンドウの制約内で情報を最適整理するためのスキル。 トークン計測、優先順位付け、圧縮、検証の一連フローを提供する。 Anchors: • High Performance Browser Networking / 適用: 計測とレイテンシ意識 / 目的: 予算内最適化 • Designing Data-Intensive Applications / 適用: リソース制約下の設計 / 目的: 効率的な構造化 • Progressive Disclosure パターン / 適用: 段階的情報開示 / 目的: 過剰読み込みの回避 Trigger: Use when optimizing context window usage, measuring token budgets, prioritizing context elements, or applying compression techniques. context window optimization, token budget, context prioritization, compression,...

Auto-router for context gathering. Detects whether the task needs codebase exploration or external knowledge research and dispatches accordingly. Trigger on "get context", "background on X", "context on X", "how does X work", architectural orientation, or any setup-before-coding request.

Attach the newest context-transfer Markdown from the current project's ./.ccb/history/ into Claude using @file. Use when the user types /continue or asks to resume with the latest transfer file in this project.

Create or update continuity ledger for state preservation across clears

Fork the current conversation to safely explore an alternative approach without losing current progress