MemCell — Watchlist | MCP100 Index

MCP100 Assessment

Watchlist / Tracking

Why we’re tracking it

Follow-up value is specific to Persistent project memory connected to coding agents over local stdio: Do scoped keyring credentials and lifecycle hooks preserve project boundaries when memories move from guest to account?

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

Best fit

Persistent project-scoped memory for coding agents across sessions and clients.

What we know

Adaptive agent-memory CLI that wires supported coding agents with non-destructive lifecycle hooks and a local MCP bridge for recall, remember, report, and document ingest.

Maintenance / activity evidence

memcell 0.10.12 published 2026-09-30T15:34:18.486Z

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

Verification gaps / watch-outs

Installation and comprehensive security have not been independently audited.
Project memory read/write plus local hook/MCP configuration. Writes: Create/update memories; Ingest files; Write `.memcell` and client integration config.
Browser approval + scoped agent credential.
Sensitive access: Project context, Agent session content, Scoped keyring credential

Persistent project memory is useful but can contain sensitive context; service retention remains unresolved.
Credentials are scoped and stored in the OS keyring.
Hooks are documented as non-destructive and fail-open.

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

All limitations / Review context

Do scoped keyring credentials and lifecycle hooks preserve project boundaries when memories move from guest to account?

The bounded review of the publisher's documented Persistent project memory connected to coding agents over local stdio and release notices identified no disclosed material condition preventing an accurately scoped public tracking entry. The resource's operational permissions remain uncertainty; this assessment does not certify installation safety.

Persistent project memory is useful but can contain sensitive context; service retention remains unresolved.
Credentials are scoped and stored in the OS keyring.
Hooks are documented as non-destructive and fail-open.
Project memory read/write plus local hook/MCP configuration.
Browser approval + scoped agent credential.
Project context, ingested documents and session reports are handled by the MemCell service; exact retention is unresolved.
Reviewed write actions: Create/update memories; Ingest files; Write `.memcell` and client integration config
Sensitive access: Project context, Agent session content, Scoped keyring credential
Installation, client compatibility and comprehensive security have not been independently audited. Authentication and access depend on the configured services.
Runtime behavior, dependencies and undisclosed vulnerabilities were not independently audited.
Documented tool authority and any remote data handling must be checked before installation or use.
Tool inventory, client matrix and permissions completeness were not requirements for this Radar tracking review.
service privacy/retention
pricing
exact successor package binding
Requires Node.js 20+.
Hooks fail open rather than blocking work.
The original memcell-ai/cli repository was archived on 2026-10-07 and points to memcell-ai/devkit and memcell-ai/mcp.
Service privacy/retention terms were not established in the reviewed official pages.
Original CLI repo archived with move notice.
Preserve repository-archive/successor caveat; do not present the archived CLI repo as the long-term implementation.
Publisher lifecycle: active package; source repository archived
Successor/context: memcell-ai/devkit and memcell-ai/mcp

Original activity observations

memcell 0.10.12 published 2026-09-30T15:34:18.486Z

Technical identity

npm/memcell

Exact implementation source: https://www.npmjs.com/package/memcell ↗
Reviewed resource type: MCP Server
Repository identity: memcell-ai/cli
Package identity: npm:memcell

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 8, 2026

Package registry ↗
memcell
Recorded package identity, distribution or download evidence.
Includes references whose attribution is not established.
Evidence source ↗
memcell.ai/docs
Source material retained for this review.
Includes references whose attribution is not established.
GitHub repository ↗
memcell-ai/cli
Repository reference; implementation attribution is not established for every document.
Includes references whose attribution is not established.

Setup · Watchlist

Install & Agents

Publisher instructions.

Source evidence observed Oct 14, 2026.

Additional setup details required — see the specific gaps below.

Warning

No runtime server process, Docker setup, OAuth or credentials were executed during this research.
Connection wizard modifies project descriptor, hooks and supported client integrations. Review changes first; OS keyring access and browser approval required.
Install documentation is not a guarantee of current package/runtime compatibility.
Publisher documents complete agent auto-connection workflow; `memcell mcp` is the stdio MCP server, distinct from package installation.
2026-10-13 targeted publisher npm README plus official repository/docs where specified; static research, not runtime test.
Never paste real credentials into a documentation example.
`memcell connect` changes local agent integration files; obtain user approval.
No live install, OAuth, paid calls, third-party mutation or credential revocation executed.
Some host removal commands are MCP100-composed based on chosen registration name, not publisher-authored.

Install / connect
Publisher instructions
Publisher-documented package/connection route; replace placeholders and verify host-specific syntax.
Publisher setup example · Publisher instructions
npm install -g memcell
Configure / authenticate
MCP100 setup guide
Use memcell connect in the intended repository to activate workspace MCP/hooks, with MemCell login/auth as required.
Use correct transport and published host config. Never paste secrets in chat; server registration is separate from software installation.
Previously reviewed configuration / prerequisites: Run memcell connect from the intended project directory; approve in browser if prompted. It auto-detects supported clients and configures agent hooks plus local MCP (`memcell mcp`). Keys go to OS keyring. Inspect client config and restart as required.
Check connection
Suggested check
memcell status checks authentication and connection health.
Remove / uninstall / revoke
MCP100 setup guide
Publisher explicitly documents `memcell hook remove` to undo repo hooks, followed by `npm uninstall -g memcell` for global CLI; local credentials are a separate optional `memcell reset` operation. Do not delete memory/project data without approval.
Local credential cache cleanup · separate from provider revocation: memcell reset clears locally cached credentials according to publisher README; do not delete user data by default.
Conditional host / package cleanup · Claude Code only · MCP100 composed guide
claude mcp remove memcell

Official Documentation & Sources

What it does

Adaptive agent-memory CLI that wires supported coding agents with non-destructive lifecycle hooks and a local MCP bridge for recall, remember, report, and document ingest.

Capability: Files & Knowledge
Resource type: MCP Server
Best fit: Persistent project-scoped memory for coding agents across sessions and clients.

Representative uses

Recall verified project context before implementation
Record directives, facts, conventions and outcomes
Ingest architecture/spec/post-mortem files
Report session outcomes and attributed savings

Limitations

Requires Node.js 20+.
Hooks fail open rather than blocking work.
The original memcell-ai/cli repository was archived on 2026-10-07 and points to memcell-ai/devkit and memcell-ai/mcp.
Service privacy/retention terms were not established in the reviewed official pages.
Original CLI repo archived with move notice.
Preserve repository-archive/successor caveat; do not present the archived CLI repo as the long-term implementation.
Publisher lifecycle: active package; source repository archived
Successor/context: memcell-ai/devkit and memcell-ai/mcp

Reviewed access

Access & risk

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

Authentication: Browser approval + scoped agent credential.
Permissions: Project memory read/write plus local hook/MCP configuration.
Data handling: Project context, ingested documents and session reports are handled by the MemCell service; exact retention is unresolved.
Cost: unknown
Risk and cautions: Persistent project memory is useful but can contain sensitive context; service retention remains unresolved.

Access scope and cautions

Write actions: Create/update memories Ingest files Write `.memcell` and client integration config
Sensitive access: Project context Agent session content Scoped keyring credential
Risk level: moderate
Security notes: Credentials are scoped and stored in the OS keyring. Hooks are documented as non-destructive and fail-open.

FAQ

Based on reviewed evidence as of 2026-10-08. Watchlist tracking is not a recommendation or a security certification.

Which tasks are documented for this implementation?

The reviewed tasks include: Recall verified project context before implementation; Record directives, facts, conventions and outcomes; Ingest architecture/spec/post-mortem files; Report session outcomes and attributed savings.

What setup and authentication are documented?

A complete connection procedure is not verified. Recorded transport: stdio. Client notes: `memcell connect` auto-detects supported agents, installs hooks/registers MCP and creates `.memcell`; `memcell status` verifies setup. These facts do not supply missing installation commands, credentials or a verified authentication procedure. Authentication: One-tap browser approval; a scoped agent credential is stored in the OS keyring.

Can it modify data or perform write operations?

Documented write actions include: Create/update memories; Ingest files; Write `.memcell` and client integration config. Recorded permission boundary: Project memory read/write plus local hook/MCP configuration.

What data-handling boundary should I consider?

Project context, ingested documents and session reports are handled by the MemCell service; exact retention is unresolved. Recorded sensitive-access context: Project context; Agent session content; Scoped keyring credential.

What limitations and risks should I check before use?

Persistent project memory is useful but can contain sensitive context; service retention remains unresolved. Requires Node.js 20+. Hooks fail open rather than blocking work. The original memcell-ai/cli repository was archived on 2026-10-07 and points to memcell-ai/devkit and memcell-ai/mcp. Service privacy/retention terms were not established in the reviewed official pages. Original CLI repo archived with move notice. Preserve repository-archive/successor caveat; do not present the archived CLI repo as the long-term implementation. Publisher lifecycle: active package; source repository archived. Successor/context: memcell-ai/devkit and memcell-ai/mcp. Credentials are scoped and stored in the OS keyring. Hooks are documented as non-destructive and fail-open. Runtime behavior, dependencies and undisclosed vulnerabilities were not independently audited. Documented tool authority and any remote data handling must be checked before installation or use.

Is this an official MCP?

OpenOri / open-cipher is listed as First-party MCP in MCP100’s current evidence.