Watchlist tracking
Reviewed for relevance and future tracking value.
MCP100 Assessment
Watchlist / Tracking
Why we’re tracking it
Follow-up value is specific to Cortex self-hosted log, transcript and incident intelligence: Do ingestion sources and shared MCP evidence preserve privacy when agent transcripts enter the operational store?
Reviewed · Documentation review; complete setup workflow not independently tested
Best fit
Self-hosted homelab/small-fleet log intelligence, incident analysis and agent-session evidence.
What we know
Native self-hosted observability/evidence system with SQLite/FTS ingestion, REST, MCP and browser workspace; npm package is a launcher for the Rust implementation.
Maintenance / activity evidence
- @dinglebear/cortex 3.18.1 published 2026-10-01T14:06:19.887Z
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.
- Read/admin actions, file-tail management, notifications, maintenance and optional LLM actions. Writes: Ingest telemetry/transcripts; Ack errors; Add/remove file-tail sources; Queue/test notifications; Backup/checkpoint/vacuum/maintenance.
- MCP/REST tokens; syslog relies on network controls.
- Sensitive access: Logs, Shell/agent activity, AI transcripts, Fleet/network topology
- Sensitive operational data store; protect tokens, DB/backups and ingestion boundaries.
Full authority, access and evidence limits are available in Review context and Access & risk.
All limitations / Review context
Do ingestion sources and shared MCP evidence preserve privacy when agent transcripts enter the operational store?
The bounded review of the publisher's documented Cortex self-hosted log, transcript and incident intelligence 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.
- Sensitive operational data store; protect tokens, DB/backups and ingestion boundaries.
- Read/admin actions, file-tail management, notifications, maintenance and optional LLM actions.
- MCP/REST tokens; syslog relies on network controls.
- Stores normalized logs/raw frames/source identities/transcripts/evidence in local SQLite.
- Reviewed write actions: Ingest telemetry/transcripts; Ack errors; Add/remove file-tail sources; Queue/test notifications; Backup/checkpoint/vacuum/maintenance
- Sensitive access: Logs, Shell/agent activity, AI transcripts, Fleet/network topology
- 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.
- Designed for trusted homelab/small fleets, not a public SIEM.
- Syslog has no app-layer auth and needs network controls.
- Launcher downloads platform-specific native binary.
- Admin MCP actions require stronger scope.
- npm package is a launcher; canonical implementation is dinglebear-ai/cortex.
Original activity observations
- @dinglebear/cortex 3.18.1 published 2026-10-01T14:06:19.887Z
Technical identity
npm/@dinglebear/cortex- Exact implementation source
- https://www.npmjs.com/package/@dinglebear/cortex ↗
- Reviewed resource type
- Host
- Repository identity
- dinglebear-ai/cortex
- Package identity
- npm:@dinglebear/cortex
- Service endpoint
- http://127.0.0.1:3100/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
- Package registry ↗
@dinglebear/cortex
Recorded package identity, distribution or download evidence.
Includes references whose attribution is not established.
- GitHub repository ↗
dinglebear-ai/cortex
Repository reference; implementation attribution is not established for every document.
Includes references whose attribution is not established.
Setup · Watchlist
Host / Desktop app setup
Publisher instructions.
Source evidence observed .
Additional setup details required — see the specific gaps below.
Download / install
Publisher instructionsOperational host with CLI, syslog/HTTP services and MCP; remote server requires separated MCP CORTEX_TOKEN and REST CORTEX_API_TOKEN.
publisher install or stdio launch · Publisher instructions
npx -y @dinglebear/cortex mcpConfigure application / integrations
MCP100 setup guideCortex publisher: `npx -y @dinglebear/cortex mcp` is the stdio MCP command, not just `npx ...` without subcommand. Daemon installation and MCP are separate. Previously reviewed client configuration / prerequisites: Launcher requires Node >=18 and downloads a native binary. Advanced setup/repair and authorization are separate operations.
Check connection
Suggested checkPublisher `cortex setup repair` for repairs; inspect MCP tools, and use `curl http://127.0.0.1:3100/health` only for daemon liveness.
Disconnect / uninstall / revoke
MCP100 setup guideFor a Claude Code registration named `dinglebear-cortex`, run `claude mcp remove dinglebear-cortex`; otherwise remove the exact entry in the chosen MCP client settings and restart/reconnect. This client-host removal command is MCP100-composed from Claude Code lifecycle documentation, not a vendor-authored uninstall. With `npx`, a separate global package uninstall normally is unnecessary. Do not delete provider data or credentials by default.
Client unregister, package uninstall, data deletion, and OAuth/API token/wallet permissions are distinct.
Remove client registration · Claude Code only · MCP100 composed guide
claude mcp remove dinglebear-cortex
What it does
Native self-hosted observability/evidence system with SQLite/FTS ingestion, REST, MCP and browser workspace; npm package is a launcher for the Rust implementation.
- Capability
- Cloud & DevOps
- Resource type
- Host
- Best fit
- Self-hosted homelab/small-fleet log intelligence, incident analysis and agent-session evidence.
Representative uses
- Ingest syslog/OTLP/Docker/file/heartbeat events
- Search/correlate incidents
- Inspect fleet topology
- Index AI-agent transcripts
- Manage notifications and maintenance
Limitations
- Designed for trusted homelab/small fleets, not a public SIEM.
- Syslog has no app-layer auth and needs network controls.
- Launcher downloads platform-specific native binary.
- Admin MCP actions require stronger scope.
- npm package is a launcher; canonical implementation is dinglebear-ai/cortex.
Reviewed access
Access & risk
Review the exact access boundary before connecting this MCP to accounts, credentials, or sensitive data.
- Authentication
- MCP/REST tokens; syslog relies on network controls.
- Permissions
- Read/admin actions, file-tail management, notifications, maintenance and optional LLM actions.
- Data handling
- Stores normalized logs/raw frames/source identities/transcripts/evidence in local SQLite.
- Cost
- self_hosted
- Risk and cautions
- Sensitive operational data store; protect tokens, DB/backups and ingestion boundaries.
Access scope and cautions
- Write actions
- Ingest telemetry/transcripts Ack errors Add/remove file-tail sources Queue/test notifications Backup/checkpoint/vacuum/maintenance
- Sensitive access
- Logs Shell/agent activity AI transcripts Fleet/network topology
- Risk level
- high
FAQ
Based on reviewed evidence as of 2026-10-08. Watchlist tracking is not a recommendation or a security certification.
Which implementation does this page cover?
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?
dinglebear-ai is listed as First-party MCP in MCP100’s current evidence.