WORKS WITH
- CCClaude Code
- CuCursor
- CxCodex CLI
- GPChatGPT
- GmGemini
One knowledge layer for every agent you run.
Stop re-explaining your project to every new session.
Every decision your team makes lives in one provider-agnostic Brain. Your agents read from it, write to it, and never start from scratch again.
DECISIONS · 47
Adopt Hono for the REST surface
decision · 2d ago
Supabase RLS on every table
infra · 5d ago
Stream extraction over batch
perf · 1w ago
SDK envelope shape frozen
api · 2w ago
Notion workspace
842 pages · imported
Linear issues
312 issues · imported
Markdown repo
50 files · importing…
STEP 1
Import what you already wrote.
Notion, Linear, a markdown repo — pull your existing project knowledge into the Brain in minutes. No re-keying, no lock-in.
$ engrym sync
✓ Connected to engrym
✓ 47 decisions indexed
✓ Watching 12 files for changes
STEP 2
Connect every tool to the same memory.
Claude Code, Cursor, Codex, ChatGPT, Gemini — point them all at one Brain over MCP. Ask once; every agent already knows.
CLAUDE.md updated
pushed to 4 agents · 3s ago
New decision extracted
“Freeze SDK envelope” · 1m ago
STEP 3
Stay in sync as the work moves.
The watcher tracks your files, extracts new decisions, and pushes context to every agent the moment it changes. Always current.
