iot-anomalies — Claude Code skill: what it does, how to install
Detects and categorizes telemetry anomalies across Cognitum Seed devices. Helpful for diagnosing erratic device behavior or vetting metrics before rolling out a firmware canary.
npx skills add ruvnet/ruflo --skill iot-anomalies -a claude-codeWhat iot-anomalies does
The skill evaluates recent device metrics using a Z-score statistical check. It invokes npx -y -p @claude-flow/[email protected] cognitum-iot anomalies DEVICE_ID using the provided target device identifier.
It sorts detected issues into explicit categories: spike, flatline, drift, oscillation, pattern-break, or cluster-outlier. When the calculated anomaly score climbs above 0.9, the agent recommends quarantining the hardware.
Diagnostics persist to agent memory through mcp__plugin_ruflo-core_ruflo__memory_store under the iot-anomalies namespace. The workflow fails immediately if npx is unavailable or if the core memory MCP tool is missing from the environment.
What it helps with
- calculate Z-scores across recent device telemetry
- classify metric anomalies into spikes, drift, flatlines, or oscillations
- verify telemetry health before advancing canary firmware rollouts
- flag erratic hardware for quarantine when anomaly score exceeds 0.9
Example prompts for the agent
- Check device
seed-dev-42for telemetry anomalies - Run metric diagnostics on
seed-891before advancing canary firmware - Classify abnormal metric readings coming from
seed-node-03
How to install iot-anomalies
1Skills CLI (Vercel)
npx skills add ruvnet/ruflo --skill iot-anomalies -a claude-code2Manually: put the files in place
~/.claude/skills/iot-anomalies/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/iot-anomalies/
1Skills CLI (Vercel)
npx skills add ruvnet/ruflo --skill iot-anomalies -a codex2Manually: put the files in place
.agents/skills/iot-anomalies/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add ruvnet/ruflo --skill iot-anomalies -a cursor2Manually: put the files in place
.cursor/skills/iot-anomalies/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install ruvnet/ruflo iot-anomalies2Manually: put the files in place
.github/skills/iot-anomalies/SKILL.md1Skills CLI (Vercel)
npx skills add ruvnet/ruflo --skill iot-anomalies -a gemini-cli2Platform CLI
gemini skills install https://github.com/ruvnet/ruflo --path plugins/ruflo-iot-cognitum/skills/iot-anomaliesManually: put the files in place
.devin/skills/iot-anomalies/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add ruvnet/ruflo --skill iot-anomalies -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about iot-anomalies
How to install iot-anomalies in Claude Code?
Run npx skills add ruvnet/ruflo --skill iot-anomalies -a claude-code. The skill file will be placed at ~/.claude/skills/iot-anomalies/SKILL.md.
What does the iot-anomalies skill do?
It runs the @claude-flow/plugin-iot-cognitum CLI tool to detect statistical anomalies in Cognitum Seed telemetry. It classifies the failure type and saves the event to MCP memory.
When does the skill recommend hardware quarantine?
Quarantine is recommended whenever the calculated anomaly score is strictly greater than 0.9.
Source text
SKILL.md · MIT
--- name: iot-anomalies description: Detect and classify telemetry anomalies on Cognitum Seed devices. Use when investigating a device that's reporting odd metrics, before approving a firmware canary advancement, or when triaging fleet-wide health alerts. allowed-tools: Bash(npx *) mcp__plugin_ruflo-core_ruflo__memory_store Read argument-hint: "<device-id>" --- Run Z-score anomaly detection on a device's recent telemetry. Steps: 1. `npx -y -p @claude-flow/[email protected] cognitum-iot anomalies DEVICE_ID` 2. Review detected anomaly types (spike, flatline, drift, oscillation, pattern-break, cluster-outlier) 3. If score > 0.9, recommend quarantine 4. Store anomaly pattern for learning: `mcp__plugin_ruflo-core_ruflo__memory_store({ key: "iot-anomaly-DEVICEID", value: "TYPE at SCORE", namespace: "iot-anomalies" })`
© ruvnet · MIT · ruvnet/ruflo
