Smriti: Shared Reasoning State for Claude Code and Codex
Git for agent reasoning state solves the multi-agent coordination collision problem.

Agent replay for Codex, but Codex is deprecated and Mubit memory feels experimental.
Teams using Codex (GitHub's now-deprecated Copilot), AI agent builders wanting shared context
Continue.dev · Github Copilot Chat
I made a CLI to sync your Codex prompts + agent reasoning + file diffs to a shared memory called Codaph (https://codaph.com). The goal was to sync agent activity across my team for a much richer understanding of the codebase.
Under the hood, it uses Mubit (https://mubit.ai) - memory engine we built based on associative retrieval. Mubit is built on the concepts of hypervectors and clustering (with time based decay for now)
Currently, Codaph works with Codex with plans to add other agentic tools soon.
Codaph is open source (vibe coded lol).
Would love to hear some feedback and if you want to try it Mubit is free (api key is available on https://console.mubit.ai)
Git for agent reasoning state solves the multi-agent coordination collision problem.
npm install for agent skills that copies prompts into Claude, Cursor, and Gemini instantly.
Turning CLI agent logs into a visual graph with sound effects is a clever way to debug thinking.
Agent skill marketplace, but cold-start problem and spam incentives unproven.
First public archive for real agent trajectories—nothing like this existed before.
Git for agent reasoning state beats orchestrators like LangGraph on parallel work.