Macrodata

Macrodata

Give Claude Code and OpenCode persistent, self-maintaining memory and autonomous scheduling.
This is experimental software. Use at your own risk.

- Layered memory - identity, journal, topics, semantic search across sessions
- Scheduling and autonomy - recurring background tasks, reminders, morning prep, self-maintenance
- Memory distillation - consolidates learnings into structured knowledge
- Dream time - overnight reflection, pattern recognition, self-improvement
- Uses your existing tools – no new APIs or third-party skills needed, but if you have them, it can use them.
- No security nightmares - runs with your existing security rules. No external APIs or third-party skills.
- Local-only - all memories stored as markdown and JSON you can read and edit.

What It Does

Learns and remembers who you are, what you're working on, and how you like to work. Analyzes your past conversations to build context. Puts working memory into every session so you never start from scratch.

Working Memory

Every session starts with context injection - your identity, current projects, daily focus, and recent activity. The agent knows who you are and what you're doing before you type anything.

State files track what matters right now:
- identity.md - how the agent should behave with you
- human.md - who you are, your preferences, your projects
- today.md - daily focus and priorities
- workspace.md - current project context
- topics - working knowledge the agent has built up

Journals

Observations, decisions, and learnings get logged to a searchable journal. Semantic search finds relevant context across all your history - journal entries, entity files, and past conversations.

Conversation Analysis

Indexes your past Claude Code and OpenCode sessions. When you're stuck on…