Opencode Claude Memory

🧠 Claude Code-compatible memory for OpenCode

Persistent, local-first shared memory for OpenCode and Claude Code — one plugin, zero migration.

This OpenCode plugin lets OpenCode read and write Claude Code-compatible Markdown memory files, so both CLIs share the same project context.

Claude Code writes memory → OpenCode reads it. OpenCode writes memory → Claude Code reads it.

Quick Start • How it works • Configuration • OpenCode 2.x • Compatibility • Migrating from v1 • FAQ

✨ At a glance

- Memory tools — memory_save / memory_delete / memory_list / memory_search / memory_read, plus the Claude Code memory instructions injected into every system prompt.
- LLM recall — before each turn a hidden agent picks the memories relevant to the query; they appear in the *first* LLM call, including single-step questions.
- Automatic extraction — after a session goes idle, a sandboxed fork reviews only the *new* part of the conversation and saves what is worth keeping. Sessions closed before the fork ran are caught up at the next start.
- Auto-dream — periodic consolidation (merge / prune / rewrite) gated on time and session count, like Claude Code.
- Claude Code-compatible — same directory, same file format, same taxonomy, same worktree handling. MEMORY.md is edited line by line so hand-organised indexes stay intact.
- Cross-platform, no shell hook — everything runs inside the OpenCode process through the plugin SDK. No python3, no jq, no wrapper.
- OpenCode 1.x and 2.x — one package serves both plugin APIs (V1 server, V2 setup) with the same options.

🚀 Quick Start

Requires OpenCode ≥ 1.18.29 (V1 plugin API) or OpenCode 2.x (@opencode/cli, tested on 2.0.22). The package exports one { id, server, setup } object: 1.x loads server, 2.x loads setup; 1.x accepts that dual object…