manage-context
コンテキストサイズを監視し、作業中に自動 compact が走らないよう事前にコンテキストを圧縮する。「コンテキストを圧縮して」「compact して」「メモリを整理」「会話を整理」「履歴を圧縮」「コンテキストが大きい」「動作が重い」などで起動。操作がない時やタスクの区切りでコンテキストが一定ラインを超えていたら自動的に /compact を実行。
Browse reusable Agent Skills, each with a clear purpose and practical guidance.
コンテキストサイズを監視し、作業中に自動 compact が走らないよう事前にコンテキストを圧縮する。「コンテキストを圧縮して」「compact して」「メモリを整理」「会話を整理」「履歴を圧縮」「コンテキストが大きい」「動作が重い」などで起動。操作がない時やタスクの区切りでコンテキストが一定ラインを超えていたら自動的に /compact を実行。
Generate Skills from HTTP MCP servers with async job patterns (submit/status/result). Use when converting MCP specifications (.mcp.json) into reusable Skills using mcp_tool_catalog.yaml, or when calling async MCP tools via JSON-RPC 2.0 with session-based polling.
MCP (Model Context Protocol) server configurations for various IDEs including setup, troubleshooting, and discovery management
MCP server development, usage, and integration patterns. Use when developing new MCP servers or integrating existing ones with agents.
Build, harden, document, and ship MCP servers to Claude standards with FastMCP 2.13+ architecture, disciplined tooling, and release-ready workflows.
Always-on MemoryManager + LearningPolicy workflow. Use when storing/retrieving memories, emitting retrieval signals, running consolidation/pruning, or when a session should default to the Atlas memory system (MemoryManager, MemoryConsolidator, LearningPolicy). Triggers: memory add/retrieve, consolidation, pruning, semantic search, or 'use the memory system by default.'
Maintain and synchronize agent templates with their associated skill templates.
The "Self-Refining" Engine. Analyzes failure patterns in other skills (e.g., `ci-autofix` failing repeatedly) and rewrites their prompts/logic to improve future performance.
Creates or updates AGENTS.md files with proper structure, skill references, and best practices. Use when adding AI assistant guidance to a project or subdirectory.