Your agents can forget.
Engram remembers.
AI coding agents restart from scratch every session. They solve the same bugs, rediscover the same conventions, and ignore what their teammates learned. Engram fixes that. One shared memory layer for every agent on your team.
Works with every MCP-compatible tool
Get started for free
Agents do not need to create a paid account themselves. A workspace owner signs up, invites teammates, and gives each agent the right member API key.
One brain. All your agents.
Across models, tools, and teammates.
Your team uses different AI tools. Claude Code here, Cursor there, Gemini or Antigravity somewhere else. They can resume their own sessions, but they have no idea what each other learned. With Engram, they share a collective brain. What one discovers, all remember.
Engram
Collective Brain
Without Engram
❌ Claude fixes a bug at 9am. Codex hits the same bug at 2pm.
❌ Your teammate's agent learns a convention. Your agent never hears about it.
❌ Agents can resume their own work, but they have no idea what teammates learned.
❌ When the conversation ends, the knowledge vanishes.
With Engram
✅ Claude fixes a bug at 9am. By 9:01, every agent on your team knows the fix.
✅ Conventions flow automatically between teammates, tools, and projects.
✅ Each session picks up where the last one left off. Knowledge compounds.
✅ The more your team works, the smarter every agent becomes.
The collective brain in action
Watch 3 different AI models learn from each other in real time. Claude discovers, Cursor validates, Gemini or Antigravity generalizes.
Not a simulation. This is how Engram works in production today.
What is MCP?
The standard that makes Engram possible.
An open standard by Anthropic
MCP (Model Context Protocol) is an open standard created by Anthropic. It lets AI agents connect to external tools and data sources through a simple, universal interface.
USB for AI
Think of it like USB for AI: one standard, every tool works. No custom integrations, no proprietary SDKs, no vendor lock-in. If your tool speaks MCP, it just works.
Engram is an MCP server
Any MCP-compatible AI tool can connect to Engram. Claude Code, Cursor, Gemini, Antigravity, Copilot, OpenCode, Windsurf, Devin. If it supports MCP, it supports Engram.
Zero integration effort
No SDK needed. No custom integration code. Just add Engram to your agent's MCP config and it gets Engram memory tools instantly.
How it connects
Your Agent
Claude, Cursor, etc.
Engram
MCP Server
Team Memory
PostgreSQL + pgvector
Why not just use config files?
CLAUDE.md, .cursorrules, AGENTS.md, .windsurfrules. Great question. Here is the honest answer.
| Feature | Config files CLAUDE.md.cursorrulesAGENTS.md | Mem0 | LangMem | Engram |
|---|---|---|---|---|
| Persists across sessions | Partial | ✓ | ✓ | ✓ |
| Shared across team | ✕ | ✓ | Partial | ✓ |
| Cross-tool (Claude + Cursor + Gemini) | ✕ | Partial | Partial | ✓ |
| Learns from outcomes | ✕ | ✕ | ✕ | ✓ |
| Confidence scoring | ✕ | ✕ | ✕ | ✓ |
| Staleness detection | ✕ | ✕ | ✕ | ✓ |
| Semantic search | ✕ | ✓ | ✓ | ✓ |
| Setup time | 0 (already there) | 5-10 min | Requires app wiring | 30 seconds (MCP config) |
| Cost | Free | Free to $249/mo | Open source / platform | Free to $99/mo |
Tip: You do not have to choose. Engram works alongside your config files (CLAUDE.md, .cursorrules, AGENTS.md). Use config files for project-specific instructions. Use Engram for team knowledge that compounds over time.
The only agent memory that gets smarter over time. Every other tool stores and retrieves. Engram stores, retrieves, and learns. When a memory helps, its confidence goes up. When it does not, it goes down. Over time, the best knowledge rises to the top.
Workspaces are your team boundary
Use one workspace for a team. Repos organize project memory inside it. Invite teammates from the dashboard so everyone gets their own key.
Frontend repo
Tools
API repo
Tools
Mobile repo
Tools
What a workspace means
- ✓The workspace is the private memory boundary for one team or company.
- ✓Repos and projects organize memories inside that boundary.
- ✓Teammates should be invited from the dashboard so each person has a separate key.
- ✓Agents in the same workspace can reuse private lessons across repos.
- ✓Different workspaces stay isolated; public memory can still enrich results.
Start ahead, not from scratch
Every new workspace starts with 100,000 source-aware public memories across agent tools, MCP, RAG/search, security, ops, engineering docs, and product/devex.
Proven approaches for common tasks. Example: "API route pattern: try-catch wrapper, validate input, check auth, then business logic." Patterns tell you HOW to do something the right way.
Team rules and coding standards. Example: "Always use Prisma migrations, never raw SQL ALTER TABLE." Conventions keep your codebase consistent.
Things that break or surprise you. Example: "async/await inside forEach does NOT work as expected. Use for...of instead." Gotchas save you hours of debugging.
Battle-tested fixes for common problems. Example: "Hydration mismatch in Next.js: use useEffect for client-only logic." Solutions give you the answer, not just the hint.
Version-specific knowledge for frameworks. Example: "Prisma 7 requires PrismaPg adapter pattern." Dependencies prevent upgrade surprises.
Step-by-step guides for complex tasks. Example: "Code Review Skill: 15 steps from pulling the branch to approving." Skills give your agent expertise on demand.
Sourced from: Covering: React, Next.js, Vue, Angular, Python, Go, Rust, TypeScript, PostgreSQL, Docker, Kubernetes, AWS, and 25+ more technologies. Plus business skills: Product Management, Sales, Marketing, HR, Legal, Finance, and more.
Other agents start empty. Yours starts with 100,000 source-aware public memories, 20 step-by-step guides, and a private workspace bank that compounds from day one.
Two memory banks. One smart agent.
Public knowledge for everyone. Private knowledge for your team.
Public Memory Bank
- ✓100,000 source-aware public memories
- ✓Agent tools, MCP, RAG/search, security, ops, and devex
- ✓20 step-by-step guides
- ✓Best-practice patterns, gotchas, solutions, and evidence checks
- ✓Updates as the public bank grows
- ✓Available to every Engram user
Private Memory Bank
- ✓Your team's proprietary knowledge
- ✓Conventions specific to your codebase
- ✓Internal architecture decisions
- ✓Incident learnings and post-mortems
- ✓Never shared outside your workspace
- ✓Isolated, encrypted, workspace-scoped
Private memory first. Public knowledge fills the gaps.
Engram searches your workspace knowledge first, then adds source-aware public practices and guides when they help.
Ready in 30 seconds
No SDK. No boilerplate. Just MCP.
Install Engram
Paste one command into your terminal. Engram writes MCP config for Claude Code, Cursor, Gemini or Antigravity, Codex, OpenCode, and VS Code/Copilot.
No manual JSON editing. No SDK. The setup command handles the config files for you.
Restart, then brief your agent
MCP clients usually load tools only when a new session starts. Restart your AI tool, then tell it what project and task it is working on.
Use normal language. Mention project, repo, role, market, stack, and the immediate task.
Start with engram_start
Your agent calls engram_start once and receives setup guidance, relevant private memories, public knowledge, and skill guides.
After that, it works normally: store durable learnings, recall before decisions, and report outcomes.
Scale it to your entire team
Solo Developer
Your Claude Code and Cursor share one brain. Switch tools without losing context.
Small Team (3-5)
Alice's agent discovers a gotcha. Bob's agent learns it. No Slack message needed. No wiki page written.
Growing Team (10-20)
Multiple repos, multiple tools. Conventions stay consistent across the org without meetings or wikis.
Enterprise (20+)
40+ agents, different models. Institutional knowledge that never walks out the door.
Your agents are already smart.
Engram makes them experienced.
Public benchmark
Measured on real memory tasks with two checks
Engram is measured two ways: first, whether it retrieves the right evidence; second, whether an LLM can answer correctly from that retrieved context.
We keep retrieval and answer accuracy separate. Both use public LoCoMo, but they measure different parts of the memory loop.
LoCoMo evidence retrieval
Full public run. Scores show whether the required source evidence appears in the retrieved memories.
| System | MRR | R@1 | R@50 | R@200 |
|---|---|---|---|---|
| Engram | 0.5345 | 40.63% | 92.19% | 97.66% |
| BM25 | 0.4676 | 34.57% | 78.19% | 85.74% |
| AgentMemory | 0.4270 | 33.20% | 79.10% | 84.96% |
| LangMem / LangGraph | 0.4002 | 28.32% | 83.98% | 94.53% |
| Mem0 OSS | 0.3999 | 28.32% | 83.92% | 94.27% |
LoCoMo answer-level judge
Codex/self-reviewed diagnostic on the same 98 tasks. Scores show whether the generated answer is correct from retrieved context.
| System | Top-50 answer | Top-200 answer | vs base model |
|---|---|---|---|
| Engram | 66.33% | 59.18% | +56.13pp |
| Session-file BM25 | 61.22% | 53.06% | +51.02pp |
| Mem0 OSS | 55.10% | 57.14% | +44.90pp |
| BM25 | 48.98% | 44.90% | +38.78pp |
| LangMem / LangGraph | 46.94% | 52.04% | +36.74pp |
| No memory | 10.20% | 9.18% | baseline |
Simple, transparent pricing
Start free. Upgrade when your agents need more memory, teammates, and API capacity.
Free
Try Engram with your personal projects.
- ✓500 private memories
- ✓1 developer
- ✓100 API calls/day
- ✓Text search + public skill packs
- ✓Community support
Developer
For solo developers shipping real products.
- ✓5,000 private memories
- ✓2 developers
- ✓2,000 API calls/day
- ✓Vector search when embeddings are configured
- ✓Usage visibility
- ✓Email support
Team
For teams sharing intelligence across agents and repos.
- ✓25,000 private memories
- ✓7 developers
- ✓10,000 API calls/day
- ✓Cross-repo memory transfer
- ✓Team dashboard
- ✓Priority support
Scale
For companies with serious AI infrastructure.
- ✓100,000 private memories
- ✓30 developers
- ✓50,000 API calls/day
- ✓SSO / SAML planning
- ✓Self-hosted option planning
- ✓SLA discussion