Package Exports

greprag
greprag/dist/index.js

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (greprag) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

GrepRAG

Episodic project memory for Claude Code and OpenCode. Auto-captures turns, compacts into hourly/daily/weekly summaries, and delivers briefings at session start. Also handles cross-project async messaging via inbox.

Quick start (both clients)

npm i -g greprag

Claude Code

Run /greprag in any Claude Code session, or:

greprag init       # set up hooks + anchor + skill

OpenCode

greprag init --opencode

This creates a project anchor in .opencode/project.json and installs the memory plugin at ~/.config/opencode/plugins/greprag-memory.js. The plugin automatically injects memory briefings at session start and stores every turn. No agent cooperation needed.

Status check (works for both):

greprag status     # check what's configured

What's active

Episodic memory: every turn captured via Stop hook, compacted into hourly/daily/weekly summaries
SessionStart recap: injects recent episodic briefings at the start of each session
Inbox: cross-project async messaging between your projects or other GrepRAG users

Commands

Command	Description
`greprag status`	Check auth, hooks, project anchor
`greprag init`	Configure hooks + anchor in current repo
`greprag init --global`	Global anchor for ephemeral sessions
`greprag doctor`	Diagnose project_id drift and orphan consolidation
`greprag inbox`	List unread messages (auto-marks read)
`greprag inbox --all`	Full message history
`greprag send "msg" --to <addr>`	Send markdown message across tenants
`greprag project-id`	Print current project's UUID

Requirements

Node.js 18+
Claude Code (the /greprag skill handles setup)
An API key (generated during email signup in the skill flow)

How it works

The CLI is a thin HTTP client. All intelligence — keyword extraction, dedup, scoring, compaction, graduation — lives server-side. The hook fires on Stop/SessionStart, calls the API, and the server handles everything else.

greprag