context▌

Context engineering curates the smallest high-signal token set for LLM tasks. The goal: maximize reasoning quality while minimizing token usage.

Establish foundational product positioning and audience context that other marketing skills reference. \n \n Captures product overview, target audience, personas, problems, competitive landscape, differentiation, objections, and brand voice in a single .agents/product-marketing-context.md document \n Offers two setup paths: auto-draft from your codebase (README, landing pages, marketing copy) for faster iteration, or conversational walkthrough for custom input \n Emphasizes capturing verbatim cu

税務法的コンテキスト（Tax Legal Context） \n このスキルは shinkoku の税務関連回答における法的・免責コンテキストを提供する。 \n 免責事項の提示 \n 免責事項が必要な回答を行う際は、 references/disclaimer.md を読み込んで以下を実行する: \n \n references/disclaimer.md の「標準免責文」を回答末尾に付記する \n 免責を強調すべきケース（グレーゾーン・高額案件等）に該当する場合は追加注意喚起を行う \n 税理士法第52条の観点から、個別具体的な税務代理行為に該当しないよう留意する \n \n 参照ファイル \n \n \n \n ファイル \n 内容 \n \n \n \n \n references/disclaimer.md \n 標準免責文・税理士法との関係・ツール制限事項・情報の正確性 \n \n \n

The agent rides the wave of peak context. When the wave crests, it commits. When it detects drift, it pulls out cleanly — saving state, handing off, and letting the next session catch the next wave.

Help a user create the Instructions field content for their Sanity Agent Context MCP. The goal is a concise set of pure deltas — only information the agent can't figure out from the auto-generated schema.

A toolkit that works with Claude Code's natural flow. Use what helps, ignore what doesn't.

Guide for implementing and maintaining context as a managed artifact alongside code, enabling consistent AI interactions and team alignment through structured project documentation.

Provides expertise in AI context management, memory architectures, and context window optimization. Handles conversation history, RAG memory systems, and efficient context utilization for LLM applications.

Context is the complete state available to a language model at inference time. It includes everything the model can attend to when generating responses: system instructions, tool definitions, retrieved documents, message history, and tool outputs. Understanding context fundamentals is prerequisite to effective context engineering.

You are tasked with understanding the intent and current state of an existing branch so you can resume or continue work effectively. The code state is your primary source of truth - it may have drifted from any associated Linear ticket.