/context-load - Load saved context
Load previously saved context to resume work or apply knowledge from past investigations.
Usage
code
/context-load [name]
Examples:
code
/context-load # Uses current directory name (e.g., ivry_web_backend) /context-load auth-refactoring # Loads specific context by name
Instructions
When this skill is invoked:
- •
Determine the context name
- •If an argument is provided, use it as the name
- •If no argument is provided, extract the directory name from the current working directory
- •Example:
/Users/asonas/ghq/github.com/ivry-inc/ivry_web_backend→ivry_web_backend
- •Example:
- •The current working directory is available in the environment information provided at the start of the conversation
- •
Search second-brain for context
- •Use
mcp__second-brain__search_memorytool with the context name as query - •Also try
mcp__second-brain__get_contextwith the topic - •Look for entries with tags containing "context" and the name
- •リポジトリ名での検索を優先: sourceフィールドにリポジトリ名が含まれているため、リポジトリ名をクエリに含めると関連度が上がる
- •例:
ivry_web_backend OpensearchIndexerWorkerのようにリポジトリ名+キーワードで検索
- •例:
- •Use
- •
Present the context to the user
- •Display the loaded context clearly
- •Highlight:
- •The original plan/decisions
- •Next steps that were identified
- •Any open questions
- •Ask if they want to proceed with the plan or need clarification
- •
If context not found
- •Suggest searching with different keywords
- •Offer to list recent memories with related tags
Notes
- •Contexts are stored in PostgreSQL (second-brain) with semantic search
- •Use tags like "context", project name, feature name for better retrieval