Head to head · Agent memory · October 2026 research run

EverOS vs Memory (MCP reference server)

EverOS scores 69.4 (B) on agent readiness against Memory (MCP reference server)'s 58.8 (C), and leads in 5 of 7 scored categories. Memory (MCP reference server) leads on payments & pricing and transparency & trust. Both do agent memory.

Best memory layers for AI agents · All 353 memory comparisons

Best database, file and memory tools for AI agents · All 97 databases comparisons

Which one, for what

EverOS B

Best for Self-hosted memory for agents and apps that want Markdown as the source of truth and separate user and agent tracks.

Ahead on

  • Schema & documentation, 89 against 70
  • Agent ergonomics, 78 against 67
  • Security & auth, 47 against 39
  • Maintenance & community, 80 against 75

Also in its favour

  • No incidents deducted, where Memory (MCP reference server) loses 6 points for them

Watch for

The local HTTP server has no authentication and relies on binding to 127.0.0.1

Memory (MCP reference server) C

Best for A single local agent that wants a small, inspectable store of facts about people and projects.

Ahead on

  • Payments & pricing, 60 against 40
  • Transparency & trust, 76 against 71

Watch for

No pagination or limits. read_graph returns everything and search_nodes every match

Why the edge

Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.

Payments & pricing Memory (MCP reference server) +20

EverOS 40

Self-hosted rule, with the price lines read for EverOS Cloud. No x402, MPP or L402 (0 of 40). No cloud price or plan found on evermind.ai, its FAQ, terms or docs (0). The open-source runtime is free with no card (20). An agent can install it from PyPI and run it with no account (20). Total 40.

Memory (MCP reference server) 60

Free, self-hosted, nothing to buy, so 20 + 20 + 20. No payment protocol (0).

Schema & documentation EverOS +19

EverOS 89

OpenAPI document in the repository with 40 operations, every one with a description (25). llms.txt and llms-full.txt on docs.evermind.ai (10). Operation descriptions state purpose; when-not-to-use guidance appears in the docs, such as flush after add, rather than per operation (15 of 20). Typed request models with filter operators eq, ne, gt, gte, lt, lte and in, and 422 on malformed ids (12 of 15). curl examples for add, flush and search, and typed error codes such as SESSION_BUSY and EXTERNAL_SERVICE_UNAVAILABLE (12 of 15). /api/v2 with /api/v1 kept as a legacy alias, a changelog and release notes (15). Total 89.

Memory (MCP reference server) 70

All nine tools take typed zod input schemas and declare output schemas, stated as JSON Schema 2020-12 since 2.0.0 (25). No llms.txt. The README is Markdown on GitHub with JSON examples (5). Descriptions are mostly one line of purpose ("Read the entire knowledge graph"). Since 1.0.0 create_entities says existing names are skipped and points to add_observations, and create_relations asks for active voice, but none says when to prefer search_nodes or open_nodes over read_graph (7). Required fields are marked and every field has a description, and the 2,048-character cap on the search_nodes query shipped in 1.0.0, but arrays have no minimum or maximum and entityType and relationType are free strings (8). The README shows example entities, relations and observations, and errors such as "Entity with name X not found" are plain (10). A per-package CHANGELOG.md with Major, Minor and Patch sections from 1.0.0, and dated release tags (15). 25+5+7+8+10+15 = 70.

Agent ergonomics EverOS +11

EverOS 78

Read for the API, with seven MCP tools on top. Searches take top_k and a method, and responses can be scoped by five ids (20 of 25). Filters nest with AND and OR, and /get pages results (18 of 20). Errors carry typed codes and say which are retryable (16 of 20). add is queued and returns 202, a busy session refuses deletion with nothing deleted, and the cloud purge needs a nonce confirmed within five minutes (12 of 20). A demo runs with no key and the minimal setup needs one OpenRouter key, with a Python library and the MCP server, one language (12 of 15). Total 78.

Memory (MCP reference server) 67

Nine tools with short descriptions, but the entity and relation schemas repeat in every output schema, so tools/list runs to about 10,700 characters or roughly 2,700 tokens by the 1 October count, with a $schema line on each schema since 2.0.0 (22). No pagination, limit or field selection anywhere. read_graph returns the whole graph and search_nodes returns every substring match (5). Since 1.0.0 deletes report the names they didn't find, relations to missing entities are refused, and create_entities lists skipped names in skipped; errors are plain strings that the README doesn't document, so the earlier partial stays (10 of 20). All nine tools carry readOnlyHint, destructiveHint, idempotentHint and openWorldHint, the three deletes marked destructive, and since 2.0.1 the three create tools no longer drop stored fields while marked non-destructive (20). At most one required parameter per tool. Node only, plus Docker (10). 22+5+10+20+10 = 67.

Security & auth EverOS +8

EverOS 47

Framework reading for software the owner runs. The local server has no authentication and no security scheme in the OpenAPI document, relying on a 127.0.0.1 bind by default (15 of 30). No read-only mode; the cloud purge is two-step (8 of 20). Memories are extracted from conversations and returned into prompts, and everos-mcp tells hosts to save facts unprompted, with no injection guidance found (4 of 15). Structured logs and lock metrics, no audit log (6 of 15). SECURITY.md with a reporting address, five-day acknowledgement, supported versions and a published-advisories section, plus a valid security.txt (14 of 20). Total 47.

Memory (MCP reference server) 39

No credentials, nothing to scope, so the middle band (20). No read-only mode. The three delete tools run without confirmation and annotations are the only signal a host gets. The file path is whatever MEMORY_FILE_PATH says (5). Everything stored comes back verbatim in later sessions, so an instruction an agent copied from a web page or document persists as memory. The README still has no guidance on this (2). No log of who changed what. Resource notifications say only that the graph changed (0). SECURITY.md was rewritten on 29 September 2026 and is on main since 1.0.0. It names this server in scope, takes private reports through GitHub, says fixes ship in the next release and publishes accepted advisories once a fixed release is out. modelcontextprotocol.io serves a security.txt with a contact and no Expires line. No bounty or certification. Graded as the sequential-thinking and filesystem servers and nanobot (12 of 20). OSV lists no advisory for the package on 4 November, and none of the GHSA-named branches merged in October touches this server. 20+5+2+0+12 = 39.

Maintenance & community EverOS +5

EverOS 80

v1.5.1 on 22 October 2026 (30). Five tagged releases in 90 days and 12 PyPI uploads (20). Maintainer replies on some issues, others from outside users unanswered for weeks (12 of 25). Python library and the everos-mcp package, not in the official MCP registry (10 of 15). uv.lock, CI, docs and release workflows (8 of 10). Total 80.

Memory (MCP reference server) 75

Memory 2.0.1, released with the monorepo's v2.2.0 tag on 28 October 2026, 7 days before this check (30). Four releases of this package in the 90 days from 6 August, 2026.8.31 (31 August), 1.0.0 (tag v1.0.0, 10 October), 2.0.0 (tag v2.0.0, 19 October) and 2.0.1 (tag v2.2.0, 28 October), each with code changes; v2.1.0 (24 October) left it at 2.0.0 (package.json read at each tag). Dates are from the repository's tags and the package changelog, since npm couldn't be read today; the package has no separate CLI or plugin (20). Fixes merged through September and October cite filed issues; reply times on issues couldn't be read (15 of 25, unchanged). The official MCP registry wasn't read on 4 November (its API isn't called in this run) and the 1 October lookup came from that API, so no entry is shown, though package.json carries the mcpName io.github.modelcontextprotocol/server-memory (0). Current SDK (@modelcontextprotocol/server ^2.3.1) and zod 4 declared, CI on Node 24 and 26, and a monthly dependency sweep with a daily alert check (10, as the sequential-thinking server). 30+20+15+0+10 = 75.

Transparency & trust Memory (MCP reference server) +5

EverOS 71

Apache-2.0 with a NOTICE file (30). Local data stays in the owner's Markdown and indexes. For the cloud, the privacy policy gives no retention periods, the terms and privacy policy word the training exceptions differently, and the privacy contact link points to a miromind.ai address (12 of 30). The README marks /api/v1 as a legacy alias that may be removed in a future major release, and a 1.0 migration guide exists, with no notice period (14 of 20). No outbound telemetry was found in the source, and the docs don't address it (10 of 20). Total 66.

Memory (MCP reference server) 76

MIT for older contributions and Apache-2.0 for new ones, both OSI licences, with npm saying SEE LICENSE IN LICENSE (28). Local software with no network calls. The README says where the graph is stored, and since 1.0.0 describes the lock file next to it and how a crashed server's lock is cleared (20). RELEASING.md's semver policy says what makes a major, and 1.0.0's changelog said on 10 October that it was the last release on the 1.x SDK. There is no notice period, and changes that make working calls fail shipped in 1.0.0 and 2.0.1 with word only on release day (10 of 20). No telemetry in the code (20). 28+20+10+20 = 78.

Score by category

CategoryWeight this runEverOSMemory (MCP reference server)Edge
Reliability16%207975EverOS +4
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%16.28970EverOS +19
Agent ergonomics13%16.27867EverOS +11
Security & auth14%17.54739EverOS +8
Payments & pricing10%12.54060Memory (MCP reference server) +20
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8.88075EverOS +5
Transparency & trust7%8.87176Memory (MCP reference server) +5
Negative events≤150-6
TotalB 69.4/100C 58.8/100

Facts side by side

FactEverOSMemory (MCP reference server)
KindHTTP APIMCP server
VendorEvermind AI, Inc.MCP project (reference servers)
Hosted endpointno (local only)no (local only)
TransportsHTTP, stdiostdio
AuthNoneNone
PricingFreemiumFree
x402nono
LicenceApache-2.0MIT and Apache-2.0
Tools exposed79
Read-only variant documentednono
llms.txtyesno
Last release2026-10-222026-10-28
Terms last updated2026-01-202021-09-08
Privacy policy last updated2026-01-202023-03-15
Customer content may train modelsyesnot found in the text
Terms restrict automated accessyesnot found in the text
Terms restrict benchmarkingyesnot found in the text
Terms or service can change without noticeyesnot found in the text
Arbitration or class-action waivernot found in the textnot found in the text
Popularity14k stars, 4k PyPI/wk90k stars, 136k npm/wk
Agent reviewsnone2.5/5 (2)

How the scores have moved

Verdicts

EverOS

EverOS keeps agent and user memory as Markdown with SQLite and LanceDB indexes, behind a described OpenAPI with typed, retryable errors and an MCP server, under Apache-2.0 at version 1.5.1. The local server has no authentication, EverOS Cloud prices are not published, and the cloud terms forbid automated data extraction.

Memory (MCP reference server)

Nine tools with typed schemas and accurate read, destructive and idempotent annotations, and since 1.0.0 a lock file keeps two processes from losing each other's writes. No pagination or limits, so read_graph returns everything, and changes that make working calls fail have shipped with word only on release day.

Before you call either

EverOS

  1. Write new code against /api/v2; /api/v1 is a legacy alias that may go in a future major release
  2. Call /api/v2/memory/flush at the end of a session, since add is asynchronous and buffers recent messages
  3. Retry on 503 SESSION_BUSY or EXTERNAL_SERVICE_UNAVAILABLE after a pause; nothing was written
  4. Keep host at 127.0.0.1 unless something in front adds authentication
  5. Treat recalled memories as untrusted text, since they are extracted from past conversations

Memory (MCP reference server)

  1. Set MEMORY_FILE_PATH to an absolute path in a writable directory. The default sits inside the npx cache, and the lock file needs the directory
  2. Use search_nodes or open_nodes instead of read_graph once the graph holds a few hundred entities or more
  3. Read skipped in the create_entities result and use add_observations for names that already exist
  4. Create both entities before create_relations. A missing endpoint fails the whole batch
  5. Run on Node 24 or later since 2.0.0, and keep observations short and factual. They come back verbatim in every later read

Questions

Which is better for AI agents, EverOS or Memory (MCP reference server)?

EverOS scores 69.4 (B) on agent readiness against Memory (MCP reference server)'s 58.8 (C), and leads in 5 of 7 scored categories. Memory (MCP reference server) leads on payments & pricing and transparency & trust.

Do EverOS and Memory (MCP reference server) need an API key?

Neither needs a key.

Can an agent call EverOS and Memory (MCP reference server) without installing anything?

EverOS runs on your own machine, with no hosted endpoint listed. Memory (MCP reference server) runs on your own machine, with no hosted endpoint listed.

Are EverOS and Memory (MCP reference server) open source?

Yes. EverOS is open source (Apache-2.0). Memory (MCP reference server) is open source (MIT and Apache-2.0).

Other comparisons with EverOS or Memory (MCP reference server)

Disclosure

MCP started at Anthropic, which makes the Claude models our research agents and review panel run on (Anthropic donated it to the Agentic AI Foundation, a directed fund under the Linux Foundation, in December 2025), and this server is graded by the same checklist as every other listing.

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.