MCP-native memory layer

Your agents forget.

AI coding agents restart from scratch every session. They solve the same bugs, rediscover the same conventions, and miss what teammates learned. Engram brings repo decisions, production gotchas, and team knowledge back to every agent.

AI tool session - Engram MCP
Human: Start Engram for housecompass.uk listing search
Agent MCP call: engram_start repo="english.cash" task="guest practice flow"

Works with every MCP-compatible tool

Claude CodeCursorCodexGeminiAntigravityCopilotOpenCodeWindsurfDevin

100,000 public memories|20 skills|200+ sources|35+ technologies|15 languages

I'm an Agent

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.

Claude Code

OpenAI Codex

Google Gemini

Engram

Collective Brain

Store

Learn

Recall

Grow

Improve

Cursor

Copilot

Windsurf

Without Engram

❌ Claude fixes a production bug in the morning. Codex hits the same bug after lunch and starts from zero.

❌ A solution learned in one repo never reaches the next repo that needs it.

❌ When context gets compacted, the agent loses why decisions were made and repeats investigation.

❌ Rules sit in config files, but nobody learns which rules actually worked.

With Engram

✅ One agent stores the fix once. The whole workspace can recall it on the next decision.

✅ Before the team rebuilds a known pattern, useful lessons flow in from other repos.

✅ After context is compacted, decisions, open loops, and gotchas return with the work.

✅ Outcome feedback teaches Engram which memories helped, so the team memory gets sharper over time.

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.

Claude Code

Cursor

Gemini

OpenAI Codex

1/5Claude Codestores a production gotcha

engram collective brain

Not a simulation. This is how Engram works in production today.

Why not just use config files?

CLAUDE.md, .cursorrules, AGENTS.md, .windsurfrules. Great question. Here is the honest answer.

Config files

CLAUDE.md.cursorrulesAGENTS.md

Persists across sessions

Config files

Partial

Mem0

✓

LangMem

✓

Engram

✓

Shared across team

Config files

Mem0

✓

LangMem

Partial

Engram

✓

Cross-tool (Claude + Cursor + Gemini)

Config files

Mem0

Partial

LangMem

Partial

Engram

✓

Learns from outcomes

Config files

Mem0

LangMem

Engram

✓

Confidence scoring

Config files

Mem0

LangMem

Engram

✓

Staleness detection

Config files

Mem0

LangMem

Engram

✓

Semantic search

Config files

Mem0

✓

LangMem

✓

Engram

✓

Setup time

Config files

0 (already there)

Mem0

5-10 min

LangMem

Requires app wiring

Engram

30 seconds (MCP config)

Cost

Config files

Free

Mem0

Free to $249/mo

LangMem

Open source / platform

Engram

Free to $99/mo

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.

Used in real projects

Agents forget. Engram brings the important context back. We saw it across six real projects.

Across six long-running projects, agents used Engram to recover prior decisions and repo-specific gotchas faster. Instead of researching the same problems again, they continued from where the work had stopped.

English.cash

★★★★★

“In long product work, Engram reduces the cost of remembering where we left off.”

Guest-practice and product-flow decisions survived across new sessions instead of being rediscovered.

Evimibul

★★★★★

“Use Engram as the starting map, then verify live screens, DB counts, and logs.”

The agent checked live data from the right place instead of retrying a path already known to be risky.

AkilConnect

★★★★★

“It answers why we chose a path without digging through commits first.”

Engram surfaced the earlier architecture decision before a scoring-order regression spread further.

Use Cases

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.

92.19%

Engram R@50

66.33%

answer accuracy from top-50 context

+8.21pp

R@50 lead over LangMem

LoCoMo evidence retrieval

Full public run. Scores show whether the required source evidence appears in the retrieved memories.

Engram92.19%

LangMem / LangGraph83.98%

Mem0 OSS83.92%

LoCoMo answer-level judge

Codex/self-reviewed diagnostic on the same 98 tasks. Scores show whether the generated answer is correct from retrieved context.

Engram66.33%

Session-file BM2561.22%

Mem0 OSS55.10%

Public benchmark

Simple, transparent pricing

Start free. Upgrade when your agents need more memory, teammates, and API capacity.

Free

$0forever

Try Engram with your personal projects.

✓500 active workspace memories
✓1 developer
✓100 agent calls/day
✓Text search + public skill packs
✓Community support

Choose Free Plan

Developer

$9/month

For solo developers shipping real products.

✓5,000 active workspace memories
✓2 developers
✓1,000 agent calls/day
✓Vector search when embeddings are configured
✓Usage visibility
✓Email support

Team

$29/month

For teams sharing intelligence across agents and repos.

✓25,000 active workspace memories
✓7 developers
✓5,000 agent calls/day
✓Cross-repo memory transfer
✓Team dashboard
✓Priority support

Scale

$99/month

For companies with serious AI infrastructure.

✓100,000 active workspace memories
✓30 developers
✓25,000 agent calls/day
✓SSO / SAML planning
✓Self-hosted option planning
✓SLA discussion

More than 30 developers? Book a call.

Your agents forget.