MCP Memory Service — Watchlist

MCP100 Assessment

Watchlist / Tracking

Why we’re tracking it

Track storage, authentication, and memory-schema changes because the service persists agent context

Reviewed Oct 5, 2026 · Documentation review; complete setup workflow not independently tested

Best fit

Persistent self-hosted agent memory with semantic search and knowledge-graph features.

What we know

A persistent memory backend exposing MCP, REST, OAuth, CLI and dashboard interfaces with local embeddings and multiple storage backends.

Maintenance / activity evidence

Observed Oct 5, 2026 (UTC)feat(metrics): opt-in Prometheus /metrics endpoint ( 1097) ( 1456)

Tracking evidence observed Oct 5, 2026. These dated signals do not establish current maintenance. Reviewed enrichment confidence: high.

Activity source · commits ↗Activity source · releases ↗

Verification gaps / watch-outs

Installation and comprehensive security have not been independently audited.
Memory store/search/admin capabilities. Writes: Store/delete memories.
OAuth for remote; optional anonymous local mode
Sensitive access: Project context, Decisions, Agent memory

Persistent memory can contain sensitive long-lived context; protect remote access and choose storage backends deliberately.

Security source · repository notice ↗Security source · policy file ↗

Full authority, access and evidence limits are available in Review context and Access & risk.

Popularity

1,987 repository stars · 8.6k package downloads / 30 days

Stars observed Oct 6, 2026. Downloads observed Oct 6, 2026.

All limitations / Review context

What storage and OAuth behavior does the current MCP memory service document?

The bounded assessment reviewed the official README, repository notices, security policy path, and security page for known material public-tracking blockers. No such blocker was identified. This is a public-tracking assessment, not an installation safety certification.

Persistent memory can contain sensitive long-lived context; protect remote access and choose storage backends deliberately.
Memory store/search/admin capabilities.
OAuth for remote; optional anonymous local mode
Memory may be local or synced to optional remote backends.
Reviewed write actions: Store/delete memories
Sensitive access: Project context, Decisions, Agent memory
Installation, client compatibility and comprehensive security have not been independently audited. Authentication and access depend on the configured services.
No independent runtime, installation, or comprehensive security audit was performed.
No installation, runtime, client matrix, or complete tool-contract audit was performed in this review.
Remote browser access requires HTTPS/OAuth.
Network exposure must be explicitly secured.

Original activity observations

2026-10-05T05:25:20Z: feat(metrics): opt-in Prometheus /metrics endpoint ( 1097) ( 1456)

Technical identity

github.com/doobidoo/mcp-memory-service

Exact implementation source: https://github.com/doobidoo/mcp-memory-service ↗
Reviewed resource type: MCP Server
Repository identity: doobidoo/mcp-memory-service
Package identity: pypi:mcp-memory-service
Service endpoint: http://localhost:8000

Evidence & Sources

Evidence basis: Recorded source evidence
Setup runtime verification: Not independently runtime-tested by MCP100 for this setup workflow.

Reviewed Oct 5, 2026 · Latest source observation Oct 6, 2026

GitHub repository ↗
doobidoo/mcp-memory-service
Recorded repository material and project history.
Evidence source ↗
pypi.org/project/mcp-memory-service/
Source material retained for this review.
Includes references whose attribution is not established.

Setup · Watchlist

Install & Agents

Claude Code

Publisher instructions with a documented client setup path.

Source evidence observed Oct 12, 2026.

Additional setup details required — see the specific gaps below.

Warning

Memory persistence and storage location should be reviewed before real data use.
Exact vendor-specific credential revoke procedure not established from checked official Setup documentation; may be external or not applicable. Do not invent.
publisher README config shape Replace the supplied credential/path placeholders privately; use only the named configuration schema.
Run the removal command in the same client and configuration scope used for registration; this does not uninstall an application, revoke third-party authorization, or delete data.
MCP100 client-scope note: this Claude Code CLI example applies only to that client and the exact registered name in the command. Desktop, Cursor and other clients use their own configuration; do not run it for a differently named entry. Run the removal command in the same client and configuration scope used for registration; this does not uninstall an application, revoke third-party authorization, or delete data.

Install / connect
Publisher instructions
Publisher-specific setup path. Follow prerequisites, then register or connect to your MCP host.
publisher documented install or connect · Publisher instructions · Review before use
```
pip install mcp-memory-service
```
Configure / authenticate
Publisher instructions
Claude Code: `claude mcp add memory -- memory server`; Claude Desktop entry uses command `memory` args `["server"]`. Local SQLite-vec needs no Cloudflare credentials.
Advanced configuration · claude desktop json · Publisher instructions
claude desktop json · Publisher instructions
{ "mcpServers": { "memory": { "args": [ "server" ], "command": "memory" } } }
Check connection
Publisher instructions
Publisher: `memory server --help`, then verify MCP `memory_search` and `memory_store` appear (do not store a test memory without consent).
read only check · Publisher instructions
memory info
Remove / uninstall / revoke
MCP100 setup guide
`claude mcp remove memory` unregisters Claude Code; optionally `pip uninstall mcp-memory-service`. Preserve saved memory DB unless explicitly deleting.
MCP100 client-scope guidance: each CLI removal example targets only its named client and exact registered entry. Package names, Desktop/Cursor configuration keys and Codex/Claude registration names may differ; use the entry you actually registered. Client removal does not revoke provider credentials.
claude mcp remove memory

Official Documentation & Sources

What it does

A persistent memory backend exposing MCP, REST, OAuth, CLI and dashboard interfaces with local embeddings and multiple storage backends.

Capability: Files & Knowledge
Resource type: MCP Server
Best fit: Persistent self-hosted agent memory with semantic search and knowledge-graph features.

Representative uses

Store/retrieve semantic memories
Share memory across sessions/agents
Use dashboard/REST/MCP access
Run local or hybrid storage

Limitations

Remote browser access requires HTTPS/OAuth.
Network exposure must be explicitly secured.

Reviewed access

Access & risk

Review the exact access boundary before connecting this MCP to accounts, credentials, or sensitive data.

Authentication: OAuth for remote; optional anonymous local mode
Permissions: Memory store/search/admin capabilities.
Data handling: Memory may be local or synced to optional remote backends.
Cost: free
Risk and cautions: Persistent memory can contain sensitive long-lived context; protect remote access and choose storage backends deliberately.

Access scope and cautions

Write actions: Store/delete memories
Sensitive access: Project context Decisions Agent memory
Risk level: moderate

FAQ

Based on reviewed evidence as of 2026-10-06. Watchlist tracking is not a recommendation or a security certification.

Which tasks are documented for this implementation?

The reviewed tasks include: Store/retrieve semantic memories; Share memory across sessions/agents; Use dashboard/REST/MCP access; Run local or hybrid storage.

What setup and authentication are documented?

Recorded installation/entrypoint step: pip install mcp-memory-service. Recorded endpoint: http://localhost:8000. Recorded transport: stdio, sse, streamable_http. Recorded environment-variable notes: MCP_ALLOW_ANONYMOUS_ACCESS; MCP_HTTP_HOST. No configuration values are supplied here. Prerequisites: Python. These are the documented setup details, not a claim that installation or client compatibility was independently tested. Authentication: OAuth 2.0 for remote MCP; anonymous local HTTP can be enabled.

Can it modify data or perform write operations?

Documented write actions include: Store/delete memories. Recorded permission boundary: Memory store/search/admin capabilities.

What data-handling boundary should I consider?

Memory may be local or synced to optional remote backends. Recorded sensitive-access context: Project context; Decisions; Agent memory.

What limitations and risks should I check before use?

Persistent memory can contain sensitive long-lived context; protect remote access and choose storage backends deliberately. Remote browser access requires HTTPS/OAuth. Network exposure must be explicitly secured. No independent runtime, installation, or comprehensive security audit was performed.

Is this an official MCP?

doobidoo is listed as Community implementation in MCP100’s current evidence.