Index agent memory with master and task-specific sub-indexes.
npx skills add https://github.com/enowdev/enowX-Skill --skill multi-brain
Please help me install this Agent Skill. Skill: multi-brain Source: https://github.com/enowdev/enowX-Skill/tree/main/skill/multi-brain Command: npx skills add https://github.com/enowdev/enowX-Skill --skill multi-brain
This skill solves the problem of fragmented agent memory, where multiple AI coding assistants lose context or repeat work because they lack a shared, persistent record of project history and decisions.
Ask the agent to run multi brain init to bootstrap the memory structure and configure your repository for shared agent context.
High-intent search queries and answers about installing and using this skill.
Persistent context sharing uses a two-level memory indexing system with master indices and task-specific sub-indexes for cross-session retrieval, preventing fragmented agent memory and redundant work across diverse AI tools.
Contextual handoff between AI assistants like Claude Code or OpenCode is achieved by reading deep context markdown files and sub-indexes, which provide a structured breadcrumb trail of previous decisions and blockers without re-reading the entire repository.
To initialize shared memory for coding agents, run the multi brain init command. This bootstraps the memory structure, configures your repository for shared agent context, and integrates non-destructively with root-level agent instruction files.
multi-brain requires non-destructive integration with root-level agent instruction files to ensure consistent memory retrieval. This approach maintains your existing configurations while adding the necessary shared memory indexing structure.
The best way to structure agent memory is a two-level indexing system: a lightweight master index for quick navigation and detailed sub-indexes for specific task areas, stored as deep context markdown files.
henrycashe26Standardize repository agent configurations for issue tracking, triage labels, and domain documentation.
aresbitSearch and analyze historical session JSONL files using jq and rg.
erclxParse .claude task, plan, and memory files to restore development context.
kjayleeManage AI agent memory across sessions with a 5-layer architecture.
