#event-sourcing

Agents you can inspect, replay and trust. An agent framework for TypeScript and Python, built on an append-only event log.

Agents you can replay, fork and trust. An agent framework for TypeScript and Python, built on an append-only event log.

Local-first operational memory and execution hub for humans and AI agents. Connectors aggregate work signals — GitHub, Slack, Microsoft 365, Box, and more — into a single local store.

A virtual world simulation platform where any Agent (LLM API / real Claude Code CLI / custom process) can be plugged in, watched on a 3D stage, commanded, and replayed. One engine, swappable world templates: debate, discussion, werewolf, aquarium, problem-solving.

Help AI agents remember, reason, and continue work across sessions through event-sourced cognitive continuity.

Local MCP event log for multi-agent workspaces

AI portfolio-management runtime for macOS. Watches obsessively, lacks feelings — the model advises, deterministic code decides. Experimental: paper trading only, not investment advice.

dsh-megamesh: super multi-agent architecture - files are messages, adoption replaces all-hands-on-deck failure, term leases replace single points, invariants replace trust, timelines replace non-reproducibility, one file serves humans and machines. Every claim carries an experiment number.

Replayable agent evidence, conflict-safe memory, and source-bound multi-agent research on flop.