The full upstream README, mirrored here for reference. Install config, tool schemas, adoption signals, and an original overview live on the Context Sync listing page.
Local-first project memory for AI coding tools over MCP.
Context Sync gives an agent a persistent memory layer for a codebase:

Context Sync is built for session continuity. Instead of relying on a model to remember earlier conversations, it stores the project context that matters and makes it retrievable through a small MCP tool surface.
The current core tools are:
set_projectrememberrecallread_filesearchstructuregitnotionInstall globally:
Auto-configuration runs during global install. After install, restart your AI tool.
If you install locally instead of globally, auto-config does not run. Use the manual config guide in docs/CONFIG.md.
Check that context-sync appears in your MCP tool list:
codex mcp listclaude mcp listIf auto-config fails, use docs/TROUBLESHOOTING.md and docs/CONFIG.md.
Run the setup wizard if you want Notion integration:
or:
Notion support is intentionally read-only at the tool surface:
notion({ action: "search", query: "..." })notion({ action: "read", pageId: "..." })Start every new session by setting the project first:
Recommended pattern:
set_project once per project/session startrecall to recover contextstructure and search to exploreread_file for focused inspectionremember when something should survive the current sessionWhen you run set_project inside a git repository, Context Sync can install hooks for automatic context capture.
Installed hooks:
post-commitpre-pushpost-mergepost-checkoutHooks are marked with Context Sync Auto-Hook, and existing hooks are backed up first.
The git tool supports:
statuscontexthotspotscouplingblameanalysisDefault locations:
~/.context-sync/data.db~/.context-sync/config.json~/.context-sync/install-status.jsonCustom database options:
context-sync --db-path /absolute/path/to/dbCONTEXT_SYNC_DB_PATHSee docs/DATA.md for storage details.