By ascorbic
Enable persistent stateful memory for Claude Code agents that automatically extracts actions, facts, and decisions from conversations, consolidates them into journals and entity files, runs nightly reflections on identity, projects, and knowledge gaps, performs end-of-day maintenance, guides setup, corrects mistakes via hooks, and provides local code reviews.
npx claudepluginhub ascorbic/macrodata --plugin macrodataExtract distilled actions and facts from today's conversations. Spawns sub-agents per conversation to avoid context blowup.
Deep nightly reflection. Self-improvement, research, pattern recognition. Runs in background with no user interaction.
End of day memory maintenance. Runs distillation, updates state files, prunes stale info. Runs in background with no user interaction.
Guide new users through macrodata setup. Creates identity, human profile, and workspace files. Use when get_context returns isFirstRun true, or user asks to set up their profile.
Give Claude Code and OpenCode persistent, self-maintaining memory and autonomous scheduling.
This is experimental software. Use at your own risk.
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.
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:
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.
Indexes your past Claude Code and OpenCode sessions. When you're stuck on something similar to before, it finds and retrieves the relevant context from previous conversations.
Periodically consolidates scattered learnings into structured knowledge. Patterns noticed across conversations become permanent understanding in your state files.
Scheduled reflection that runs while you're away. Reviews recent activity, notices patterns, updates state files, and prepares for tomorrow. Researches best practices. The agent maintains itself.
Some autonomous agent systems run their own shell, execute third-party skills, and expose APIs - creating prompt injection vectors, credential leaks, and remote code execution risks.
Macrodata runs inside Claude Code's existing permission model. It uses only the tools you've already installed and approved. No external APIs, no third-party skill downloads, no new attack surface. Scheduled tasks run through the same Claude Code instance with the same permissions you've already granted.
The daemon is a simple cron runner that spawns Claude Code when reminders fire. All state is local markdown files. Nothing phones home.
/plugin marketplace add ascorbic/macrodata
/plugin install macrodata@macrodata
~/.config/opencode/opencode.json:
{
"plugin": ["@macrodata/opencode"]
}
Launch the app and ask to set up Macrodata.
The layered memory architecture is inspired by Letta, and particularly the Void bot by Cameron Pfiffer. The ambient compute and dream time concepts are inspired by Strix by Tim Kellogg. I first explored these ideas in my Acme agent, that I built in December 2025, and which continues to run and evolve on my Mac Mini. Acme built Innie, my first work agent, and Macrodata is these tools and ideas made available as a plugin for Claude Code and OpenCode.
MIT
Team-oriented workflow plugin with role agents, 27 specialist agents, ECC-inspired commands, layered rules, and hooks skeleton.
Admin access level
Server config contains admin-level keywords
Complete collection of battle-tested Claude Code configs from an Anthropic hackathon winner - agents, skills, hooks, rules, and legacy command shims evolved over 10+ months of intensive daily use
Complete collection of battle-tested Claude Code configs agents, skills, hooks, rules, and legacy command shims evolved over 10+ months of intensive daily use
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer
AI-supervised issue tracker for coding workflows. Manage tasks, discover work, and maintain context with simple CLI commands.