Product

Everything in the box

One install gets you the engine, the MCP server, the dashboard, and the Python library. Here's exactly what's inside, how to install it, and what each tier unlocks.

The Knowledge Graph

Your agent's memory, as a living map you can grab

Every memory becomes a linked entity. Open the Graph tab and the whole knowledge base lays itself out in front of you — hundreds of nodes you can reach in and move.

127.0.0.1:8700 / knowledge-graph
Engraphis Knowledge Graph tab: 345 memories rendered as an interactive force-directed graph, with live physics controls and a top-connected entity ranking
Fast

Real-time physics, hundreds of nodes

A force-directed layout runs live in your browser. Grab a node and the graph re-settles instantly — 345 memories here, no lag and no round-trip to a server.

Interactive

A tool, not a picture

Drag nodes to pull threads apart; tune repel force, link distance and gravity; search for any entity; then Refresh, Reheat, or Fit. Hide unconnected noise or reveal every link label.

Interconnected

See the shape of what it knows

Related memories link together and the most-connected concepts rise to the top, so the themes your agent keeps returning to are obvious at a glance.

Install

Pick the install that matches what you need

$ pip install "engraphis[all]" # dashboard + MCP server + code graph + everything $ pip install "engraphis[server]" # dashboard + REST API $ pip install "engraphis[mcp]" # MCP server only $ pip install engraphis # core library, numpy only, fully offline

First run downloads all-MiniLM-L6-v2 (~80 MB). Without it, the engine falls back to a deterministic offline embedder. It always runs.

Quickstart

Three ways to run it

Dashboard WebUI

The headline experience

$ pip install "engraphis[server]" $ engraphis-dashboard # 127.0.0.1:8700
MCP server

Claude Code, Cursor, Zed

$ pip install "engraphis[mcp]" $ engraphis-init $ claude mcp add engraphis -- engraphis-mcp
Python library

Same service, no server

from engraphis.service import MemoryService mem = MemoryService.create("engraphis.db") mem.remember("Auth migrated to PASETO.", workspace="acme", repo="api")
MCP tools

18 tools, once your agent connects

engraphis_remember and engraphis_recall, scoped to workspace, repo, and session, plus governance, code awareness, and session lifecycle.

CategoryToolWhat it does
Writeengraphis_rememberStore a fact; deterministically resolved (add / reinforce / supersede)
Writeengraphis_record_eventAppend a lightweight episodic log entry
Writeengraphis_linkExplicitly connect two related memories
Readengraphis_recallHybrid vector + lexical + graph recall
Readengraphis_recall_groundedCited answer from retrieved memories, or abstain
Readengraphis_recall_proactive"What should I know right now," no query needed
Readengraphis_whyCurrent answer, plus what it superseded
Readengraphis_timelineFull bi-temporal history, oldest first
Codeengraphis_index_repoParse a repo into the code symbol graph
Codeengraphis_search_codeFind symbols by name, with callers
Governanceengraphis_forgetRetire a memory: bi-temporal close, never deleted
Governanceengraphis_pinExempt from future automatic decay / pruning
Governanceengraphis_correctReplace content without losing history
Sessionengraphis_start_session / _end_sessionSession lifecycle with cross-session handoff
Opsengraphis_statsMemory counts for health checks
Configuration

Set entirely through environment variables

Set these directly or drop them in a .env file. See .env.example in the repo for the full list.

Env varDefaultDescription
ENGRAPHIS_DB_PATH./engraphis.dbSQLite database file
ENGRAPHIS_HOST127.0.0.1Server bind address
ENGRAPHIS_PORT8700Dashboard port
ENGRAPHIS_API_TOKENIf set, REST API requires a bearer token
ENGRAPHIS_EMBED_MODELall-MiniLM-L6-v2sentence-transformers model
ENGRAPHIS_LLM_PROVIDERopenaiopenai | anthropic | google | openrouter
ENGRAPHIS_LLM_API_KEYLLM API key (only for optional chat/synthesis)
ENGRAPHIS_LICENSE_KEYPro/Team key (or ~/.engraphis/license.key)
Comparison

How Engraphis stacks up

Engraphis owns self-hosted, coding-agent-native, and no per-operation metering. Where a competitor is the better pick, we say so.

vs. mem0

mem0 is an Apache-2.0 memory library with a hosted SaaS on top: a genuinely good product. The difference is the deployment model and where the paywall falls. mem0 gates graph memory behind a paid tier; Engraphis keeps the whole recall engine free.

mem0Engraphis
Default deploymentHosted-first SaaSSelf-hosted-first, local by default
Billed per memory / retrievalYes, hostedNo, runs on your box
Graph memoryGated to a paid tierIn the free core
Coding-agent integrationSDK glueNative MCP server (18 tools)
Code symbol graphNoYes

vs. Zep

Zep is a cloud, credit-metered memory service that deprecated its self-hosted Community Edition, vacating the on-prem/private niche entirely. Any team disqualified from a hosted memory cloud is a customer Zep has walked away from.

ZepEngraphis
DeploymentCloud only, self-hosted CE deprecatedSelf-hosted, first-class
Billing modelCredit-metered per operationNo metering, flat license
Offline / air-gappedNoYes
On-prem / regulated / residency-boundNo longer servedThe core use case

vs. Letta (MemGPT)

Letta is an open-source agent framework with memory built in, and its free tier is genuinely generous. The real difference is scope: Letta asks you to build on it; Engraphis asks you to add it.

LettaEngraphis
What it isAgent framework with memory insideDrop-in memory layer, bring your own agent
Adoption costAdopt the frameworkAdd one MCP server, keep your stack
Coding-agent-nativeVia the frameworkNative MCP, no framework buy-in
Code symbol graphNoYes
Bi-temporal truth + supersession viewLimitedFirst-class + Inspector diffs

Competitor pricing and features change. Verify current figures against each vendor's live pricing page.

Pricing

Free core, forever. Flat per-seat for the rest.

A hosted memory cloud meters you per memory and per retrieval. Engraphis is a flat license for software that runs on your box — the engine is free, and you pay only when memory becomes something a team needs to observe, prove, and govern. Never by the lookup.

Community

Free

$0
Apache-2.0, available now

Best for a solo developer

  • The complete recall engine — unlimited, on your machine
  • MCP server, dashboard with Memory Inspector, CLI, library
  • Yours under Apache-2.0, free forever
Install free
Team

Team

$20 / seat / mo
or $200 / seat / yr — 2 months free

Best for a team sharing one memory

  • Everything in Pro
  • Multi-user Inspector — logins, roles, seats
  • Shareable analytics HTML reports
  • SSO/RBAC, audit exports, SLA
Start 3-day free trial
Enterprise (custom): BYOC, encryption-at-rest, security review, dedicated support. Contact us to talk scope.

License keys are verified offline (Ed25519-signed), no phone-home. Pro and Team start with a 3-day free trial. Community is free forever.

Give your agents a memory.

> free and open-source at the core, runs entirely on your machine