using-second-brain

What problem does it solve?

It prevents the assistant from answering without first loading the persona’s identity and the most relevant Second-Brain memory context, especially when the user references a specific wiki slug or entity.

Core Features & Use Cases

• Persona-first session setup: Uses the persona-context content already injected at session start (identity card, top wiki hits, and installed plugin catalog) to anchor all subsequent responses without restating the infrastructure.
• Hard wiki lookup rule for named slugs: If the user’s message names a host/service/project/decision/blocker/wiki page slug, it mandates a mcp__knowledge-base__knowledge_search call on that literal name before answering.
• Context-aware, low-friction responses: Surfaces relevant prior decisions briefly (one sentence each), avoids unnecessary process narration, and uses an engagement gate to reduce flow disruption.
• Graph edge safety before graph-affecting actions: When edits/refactors/deletes/renames map to a knowledge/graph node, it checks dependencies via mcp__knowledge-base__knowledge_neighbors to warn once about costly-to-miss edges.
• Native-memory conflict handling: Flags and resolves conflicts when Claude Code’s built-in auto-memory contradicts the second-brain graph/wiki on load-bearing facts.

Quick Start

Ask for help using second-brain-aware responses, and if you reference a specific wiki slug or project name, ensure the assistant queries the knowledge base on that name before giving your answer.