Open-source observability platform with S3 storage

Ingest and explore logs, traces, and metrics stored in your S3 buckets. Query with natural language. Create AI agents that detect anomalies and send daily/weekly reports to your inbox.

Website • Playground • Discord • Twitter • Documentation

What is Monoscope? • Cloud vs Self-hosted • Quick Start • AI Agents • Roadmap

What is Monoscope?

Monoscope is an open-source observability platform that stores your telemetry data in S3-compatible storage. Self-host it or use our cloud offering.

Core capabilities:

• 💰 S3 storage — Store years of logs, metrics, and traces affordably in your own S3 buckets
• 💬 Natural language queries — Search your data using plain English via LLMs
• 🤖 AI agents — Create agents that run on a schedule to detect anomalies and surface insights
• 📧 Email reports — Receive daily/weekly summaries of important events and anomalies
• 🔭 OpenTelemetry native — 750+ integrations out of the box
• ⚡ Live tail — Stream logs and traces in real-time
• 🕵️ Unified view — Correlate logs, metrics, traces, and session replays in one place

Cloud vs Self-hosted

In both options, you bring your own S3 buckets—your data stays yours.

→ Start free on Cloud or continue below to self-host.

Quick Start

git clone https://github.com/monoscope-tech/monoscope.git
cd monoscope
docker-compose up

Visit http://localhost:8080 (default: admin/changeme)

Send Test Data

Populate your dashboard with test telemetry:

monoscope auth login

# Single event — see your message appear in the dashboard
monoscope send-event -m "Hello from Monoscope"

# Sustained load — stress-test the pipeline
monoscope telemetrygen --kind=trace --rate=5 --count=50

CLI

Manage your Monoscope project from the terminal — search logs, query metrics, manage monitors and dashboards, triage issues, and more.

curl https://monoscope.tech/install.sh | sh
monoscope auth login

See the CLI reference for the full command list.

Agentic pipeline

Every CLI command emits a stable JSON envelope, so you can chain discovery → search → triage without manual munging. This is the same pipeline a Claude Code skill runs end-to-end:

# 1. What services exist? (Precomputed; no aggregation query needed.)
SVC=$(monoscope facets resource.service.name --top 1 \
        | jq -r '.["resource.service.name"][0].value')

# 2. Grab one error event from that service — id only, ready to chain.
ID=$(monoscope logs search 'severity.severity_text=="error"' \
        --service "$SVC" --first --id-only)

# 3. Pull the surrounding 5 minutes of traffic, with a per-trace summary
#    showing which other services were affected.
monoscope events context --window 5m --summary \
  --at "$(monoscope events get "$ID" | jq -r .timestamp)" \
  | jq '.traces | sort_by(-.error_count) | .[0:3]'

# 4. Acknowledge the open issue once you have a hypothesis.
monoscope issues list --service "$SVC" --status open \
  | jq -r '.data[].id' | head -3 \
  | xargs -I {} monoscope issues ack {}

Each step's output shape is documented and stable:

Set MONOSCOPE_AGENT_MODE=1 (or run with --agent) to force JSON output and disable interactive prompts — auto-detected when CI or CLAUDE_CODE is set.

Claude Code Skills

Let Claude investigate incidents, triage alerts, and write KQL queries using the monoscope CLI — install the skills plugin:

# Claude Code
claude plugin marketplace add monoscope-tech/skills
claude plugin install monoscope-skills@monoscope-skills

# or via npx (Cursor, Cline, Copilot, and other agents)
npx skills add monoscope-tech/skills

Restart Claude Code after installation. Skills activate automatically when relevant — e.g. "investigate the 500 errors in payment-api" or "do an on-call sweep".

See github.com/monoscope-tech/skills for the full skill list and documentation.

MCP Server

Monoscope exposes itself as a Model Context Protocol server at /api/v1/mcp, so any MCP-aware client (Claude Desktop, Cursor, Cline, custom agents) can search events, manage monitors and dashboards, triage issues, and run composite workflows like search_events_nl (natural-language → KQL) and analyze_issue (LLM-assisted root-cause).

{
  "mcpServers": {
    "monoscope": {
      "url": "https://api.monoscope.tech/api/v1/mcp",
      "headers": { "Authorization": "Bearer YOUR_API_KEY" }
    }
  }
}

Every public REST endpoint is auto-registered as a verb-first tool (list_monitors, search_events, mute_monitor, …). See the MCP reference for the full protocol, tool catalog, and examples.

Integration

Auto-instrument your apps

pip install opentelemetry-distro opentelemetry-exporter-otlp
opentelemetry-bootstrap -a install
OTEL_SERVICE_NAME="my-app" \
OTEL_EXPORTER_OTLP_ENDPOINT="http://localhost:4317" \
opentelemetry-instrument python myapp.py

npm install --save @opentelemetry/auto-instrumentations-node
OTEL_SERVICE_NAME="my-app" \
OTEL_EXPORTER_OTLP_ENDPOINT="http://localhost:4317" \
node --require @opentelemetry/auto-instrumentations-node/register app.js

curl -L https://github.com/open-telemetry/opentelemetry-java-instrumentation/releases/latest/download/opentelemetry-javaagent.jar -o otel-agent.jar
OTEL_SERVICE_NAME="my-app" \
OTEL_EXPORTER_OTLP_ENDPOINT="http://localhost:4317" \
java -javaagent:otel-agent.jar -jar myapp.jar

# Install OpenTelemetry Operator
kubectl apply -f https://github.com/open-telemetry/opentelemetry-operator/releases/latest/download/opentelemetry-operator.yaml

# Configure auto-instrumentation
kubectl apply -f - <<EOF
apiVersion: opentelemetry.io/v1alpha1
kind: Instrumentation
metadata:
  name: my-instrumentation
spec:
  exporter:
    endpoint: http://monoscope:4317
  propagators:
    - tracecontext
    - baggage
EOF

# Annotate your deployments for auto-instrumentation
kubectl patch deployment my-app -p \
  '{"spec":{"template":{"metadata":{"annotations":{"instrumentation.opentelemetry.io/inject-java":"my-instrumentation"}}}}}'

Natural Language Queries

Query your telemetry data in plain English:

• "Show me all errors in the payment service in the last hour"
• "What caused the spike in response time yesterday?"
• "Which endpoints have the highest p99 latency?"

AI Agents & Reports

Create AI agents that monitor your systems on a schedule:

• Scheduled analysis — Agents run at intervals you define (hourly, daily, weekly)
• **Anoma