Watchlist tracking
Reviewed for relevance and future tracking value.
MCP100 Assessment
Watchlist / Tracking
Why we’re tracking it
Track storage and retrieval behavior as the memory implementation changes
Reviewed · Documentation review; complete setup workflow not independently tested
Best fit
Shared persistent coding-agent memory across MCP clients
What we know
Local-first shared memory layer for coding agents with stdio/HTTP MCP, project-scoped memory, dashboards and multi-agent coordination modes.
Maintenance / activity evidence
- Observed (UTC)chore(release): prepare 1.9.9
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.
- Memory operations vary by selected tool profile. Writes: Store/update/import memories; Team coordination state.
- Authentication not yet verified.
- Sensitive access: Long-lived coding/project context
- Persistent memory can contain proprietary context; HTTP exposure and tool profile should be chosen deliberately.
Full authority, access and evidence limits are available in Review context and Access & risk.
Popularity
807 repository stars
Stars observed Sep 25, 2026.
Preserve any already-reviewed adoption metrics in MCP100. Batch 4 did not refresh package downloads because the package-stats provider had insufficient credits.
Preserve any already-reviewed adoption metrics in MCP100. Batch 4 did not refresh package downloads because the package-stats provider had insufficient credits.
All limitations / Review context
Which memory backends and MCP operations are currently documented?
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 proprietary context; HTTP exposure and tool profile should be chosen deliberately.
- Memory operations vary by selected tool profile.
- Local-first memory; optional background/VPS/container deployments can expose it over HTTP.
- Reviewed write actions: Store/update/import memories; Team coordination state
- Sensitive access: Long-lived coding/project context
- Installation, client compatibility and comprehensive security have not been independently audited. Authentication and access depend on the configured services.
- A project-specific SECURITY.md policy was not available at the fetched official path; the review remains limited to published repository notices.
- No installation, runtime, client matrix, or complete tool-contract audit was performed in this review.
- The fetched official SECURITY.md path returned HTTP 404; it was not treated as positive evidence.
- Requires Node.js 22.18+.
- HTTP/background service should be exposed only intentionally.
Original activity observations
- 2026-10-02T10:03:55Z: chore(release): prepare 1.9.9
Technical identity
github.com/avids2/memorix- Exact implementation source
- https://github.com/avids2/memorix ↗
- Reviewed resource type
- MCP Server
- Repository identity
- avids2/memorix
- Package identity
- npm:memorix
- Service endpoint
- http://localhost:3211/mcp
Evidence & Sources
- Evidence basis
- Recorded source evidence
- Setup runtime verification
- Not independently runtime-tested by MCP100 for this setup workflow.
Reviewed · Latest source observation
- GitHub repository ↗
avids2/memorix
Recorded repository material and project history.
Includes references whose attribution is not established.
Setup · Watchlist
Install & Agents
Codex
Publisher instructions with a documented client setup path.
Source evidence observed .
Additional setup details required — see the specific gaps below.
Install / connect
Publisher instructionsRequires Node.js >=22.18 and Git. Install Memorix, then run the official memorix setup --agent <agent> --global for the actual client; Codex now uses a personal marketplace/plugin path rather than assuming a direct config.toml registration.
npm install -g memorixmemorix setup --agent codex --globalConfigure / authenticate
Publisher instructionsChoose a supported target agent and run memorix setup --agent codex --global (or selected agent). It manages plugin/config/hooks differently for each target; use memorix setup --list to see supported clients and memorix doctor agents --agent codex to verify its managed integration.
Check connection
MCP100 setup guideUse memorix doctor agents --agent codex to inspect Codex plugin/agent integration and unresolved approvals. The publisher doctor checks plugin/hook status rather than proving a memory write; also verify the actual tools in the client.
memorix setup --listmemorix doctor agents --agent codexRemove / uninstall / revoke
Publisher instructionsPreview `memorix uninstall --dry-run`; run `memorix uninstall --background --hooks` to stop service/remove hooks. Manually remove any remaining MCP client entries reported by the command. `npm uninstall -g memorix` removes global CLI. NEVER use `--purge-data` for normal disconnect: it deletes local memories.
memorix uninstall --dry-runAdvanced configuration · stop and remove hooks
stop and remove hooks · Review before use
memorix uninstall --background --hooksAdvanced configuration · conditional global package uninstall
conditional global package uninstall · Review before use
npm uninstall -g memorixAdvanced configuration · optional destructive data purge
optional destructive data purge · Review before use
memorix uninstall --yes --background --hooks --purge-data
What it does
Local-first shared memory layer for coding agents with stdio/HTTP MCP, project-scoped memory, dashboards and multi-agent coordination modes.
- Capability
- Files & Knowledge
- Resource type
- MCP Server
- Industry
- Software & Technology
- Best fit
- Shared persistent coding-agent memory across MCP clients
Representative uses
- Install one global agent-memory integration
- Search/store project memories
- Run shared HTTP memory service
- Use micro/lite/team/full tool profiles
Limitations
- Requires Node.js 22.18+.
- HTTP/background service should be exposed only intentionally.
Reviewed access
Access & risk
Review the exact access boundary before connecting this MCP to accounts, credentials, or sensitive data.
- Authentication
- Not documented in the reviewed sources
- Permissions
- Memory operations vary by selected tool profile.
- Data handling
- Local-first memory; optional background/VPS/container deployments can expose it over HTTP.
- Cost
- free
- Risk and cautions
- Persistent memory can contain proprietary context; HTTP exposure and tool profile should be chosen deliberately.
Access scope and cautions
- Write actions
- Store/update/import memories Team coordination state
- Sensitive access
- Long-lived coding/project context
- 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?
What setup and authentication are documented?
Can it modify data or perform write operations?
What data-handling boundary should I consider?
What limitations and risks should I check before use?
Is this an official MCP?
AVIDS2 is listed as Community implementation in MCP100’s current evidence.