UniFi semantic-analysis MCP — fleet health, anomaly detection, cross-site analytics
Config is the same across clients — only the file and path differ.
{
"mcpServers": {
"unifi": {
"env": {
"UNIFI_API_KEY": "<your-key>",
"UNIFI_API_KEY_OWNER": "<owner-key-or-same-key-if-role=owner>"
},
"args": [
"-y",
"@us-all/unifi-mcp"
],
"command": "npx"
}
}
}Are you the author?
Add this badge to your README to show your security score and help users find safe servers.
The MSP-style UniFi MCP — built around the official Site Manager API + Cloud Connector with cross-site analytics no other UniFi MCP exposes. > 54 tools split across 7 semantic-analysis aggregations, 9 raw Site Manager, and 35 Cloud Connector — plus 2 optional local controller tools that surface per-port error counters and SFP DDM the Cloud API doesn't expose. Severity verdicts (healthy/info/warning/critical) on top of curated thresholds. 8 MCP Prompts (4 fleet-wide ops + 4 MSP workflows). Read
Run this in your terminal to verify the server starts. Then let us know if it worked — your result helps other developers.
npx -y '@us-all/unifi-mcp' 2>&1 | head -1 && echo "✓ Server started successfully"
After testing, let us know if it worked:
Five weighted categories — click any category to see the underlying evidence.
No known CVEs.
Checked @us-all/unifi-mcp against OSV.dev.
Click any tool to inspect its schema.
triage-site-degradationFleet ops: triage site degradation with device statuses, WAN uptime, reboots, and bandwidth analysis
firmware-rollout-auditFleet ops: audit firmware versions across fleet and plan rollouts
wan-uptime-reportFleet ops: generate WAN uptime reports with outage analysis
cross-site-anomaly-detectionFleet ops: detect anomalies across sites by comparing clients, traffic patterns, and firmware
msp-onboard-site-checklistMSP workflow: onboarding checklist for new sites
msp-monthly-client-reportMSP workflow: generate monthly client activity and health reports
msp-fleet-firmware-planMSP workflow: plan firmware updates across entire managed fleet
msp-bandwidth-complaint-investigationMSP workflow: investigate and resolve bandwidth complaints
Be the first to review
Have you used this server?
Share your experience — it helps other developers decide.
Sign in to write a review.
Others in other
Pi Coding Agent extension (CLI-first) — routes bash/read/grep/find/ls through lean-ctx CLI for strong token savings. Optional MCP bridge can register advanced tools.
Autonomous spec-to-product coding-agent CLI with an MCP server exposing 34 tools over stdio.
97% token reduction for AI coding sessions — zero deps, 21 languages, MCP server
App framework, testing framework, and inspector for MCP Apps.
MCP Security Weekly
Get CVE alerts and security updates for io.github.us-all/unifi and similar servers.
Start a conversation
Ask a question, share a tip, or report an issue.
Sign in to join the discussion.
The MSP-style UniFi MCP — built around the official Site Manager API + Cloud Connector with cross-site analytics no other UniFi MCP exposes.
54 tools split across 7 semantic-analysis aggregations, 9 raw Site Manager, and 35 Cloud Connector — plus 2 optional local controller tools that surface per-port error counters and SFP DDM the Cloud API doesn't expose. Severity verdicts (
healthy/info/warning/critical) on top of curated thresholds. 8 MCP Prompts (4 fleet-wide ops + 4 MSP workflows). Read-only — Ubiquiti's API keys don't ship write yet.
npx -y @us-all/unifi-mcp --doctor
Validates env vars, pings Site Manager API, probes Cloud Connector (if owner key set), and checks category toggles before starting. Exits non-zero on critical issues so it works in CI / pre-deploy scripts.
site-health-timeline, summarize-site, firmware-inventory, compare-sites, wan-uptime-trend, top-clients-by-bandwidth, list-sites-overview. No other UniFi MCP exposes these.healthy / info / warning / critical / unknown with a curated reason. Curated thresholds (e.g. WAN uptime <90% = critical, startupTime <1h = critical post-reboot)./v1/connector/consoles/{id}/... proxy. connectorAvailable (capability) vs connectorResolved (this-call) split.caveats array surfacing partial failures (e.g. Site Manager API can't window-bound WAN uptime — that's surfaced explicitly).triage-site-degradation, firmware-rollout-audit, wan-uptime-report, cross-site-anomaly-detection. MSP workflows: msp-onboard-site-checklist, msp-monthly-client-report, msp-fleet-firmware-plan, msp-bandwidth-complaint-investigation.@us-all/* MCPs (default ~5K tokens with owner key). Fleet of 200+ devices analyzable inside a single session.summarize-site renders as a fleet-status card on ChatGPT clients (online %, WAN uptime, gateway, devices) via _meta["openai/outputTemplate"]. Claude clients receive the same JSON content.MCP_TRANSPORT=http for ChatGPT Apps SDK or remote clients (Bearer auth via MCP_HTTP_TOKEN).UNIFI_LOCAL_* env enables 2 tools that bypass the Cloud Connector and hit the controller's legacy /api/s/{site}/stat/device/{mac} directly on the LAN: get-port-errors (port-level rx/tx errors, link-flap counters, SFP DDM — Rx/Tx Power dBm, temperature, voltage, TX/RX fault) and list-port-flap-summary (fleet-wide port instability ranking). Surfaces data the Integration API doesn't expose. Requires LAN reachability.Connect the server to Claude Desktop or Claude Code, then paste any of these:
healthy with severity, top 3 issues."