hot-shots

Node.js client for StatsD and DogStatsD (Datadog's extended StatsD protocol). Sends metrics (counters, gauges, timers, histograms, sets, distributions) and service checks to a StatsD server or Datadog Agent via UDP/UDS. Supports Datadog extensions including tags, events, service checks, and DogStatsD socket. Maintained by Brightcove.

Evaluated Mar 06, 2026 (0d ago) v10.x
Homepage ↗ Repo ↗ Developer Tools statsd datadog metrics monitoring observability node.js udp dogstatsd
⚙ Agent Friendliness
64
/ 100
Can an agent use this?
🔒 Security
82
/ 100
Is it safe for agents?
⚡ Reliability
83
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
--
Documentation
85
Error Messages
75
Auth Simplicity
100
Rate Limits
88

🔒 Security

TLS Enforcement
80
Auth Strength
78
Scope Granularity
78
Dep. Hygiene
85
Secret Handling
90

UDP transmission is unencrypted — use Unix Domain Socket (UDS) mode for local agent communication. No credentials in client. Metric data may include sensitive cardinality (user counts, error rates) — treat as internal telemetry.

⚡ Reliability

Uptime/SLA
85
Version Stability
88
Breaking Changes
85
Error Recovery
75
AF Security Reliability

Best When

Your team uses Datadog or a StatsD-compatible metrics backend and you need a lightweight, battle-tested Node.js metrics client with full DogStatsD extension support.

Avoid When

You're using Prometheus or need OpenTelemetry compatibility — use prom-client or @opentelemetry/api for those ecosystems.

Use Cases

  • Instrument agent pipelines with custom metrics (task counts, latency, error rates) sent to Datadog or StatsD
  • Track LLM API call counts, token usage, and response times as gauge/histogram metrics in observability platforms
  • Send heartbeat checks and service status events to Datadog from agent health monitoring loops
  • Measure agent task queue depth and processing throughput using StatsD counters and gauges
  • Tag metrics with agent instance identifiers, model names, and workflow IDs for fine-grained observability

Not For

  • Prometheus metrics — hot-shots sends to StatsD/Datadog only; use prom-client for Prometheus exposition format
  • Log aggregation — hot-shots is for numeric metrics, not structured log shipping
  • Applications not using Datadog or StatsD — use OpenTelemetry for vendor-neutral observability

Interface

REST API
No
GraphQL
No
gRPC
No
MCP Server
No
SDK
Yes
Webhooks
No

Authentication

Methods: none
OAuth: No Scopes: No

Authentication to the metrics backend is handled at the StatsD/Datadog Agent level. hot-shots sends to localhost agent via UDP — no auth in the client itself.

Pricing

Model: open_source
Free tier: Yes
Requires CC: No

MIT license. Metrics backend costs (Datadog) are separate. hot-shots client itself is free.

Agent Metadata

Pagination
none
Idempotent
Full
Retry Guidance
Not documented

Known Gotchas

  • UDP is fire-and-forget — metrics may be dropped silently if the StatsD agent is down or the UDP buffer is full; always add an errorHandler to detect drops
  • Metric names use dot notation by default — Datadog converts dots to underscores in the UI; use underscores in metric names for consistency
  • Tags must be strings in 'key:value' format — passing objects or non-string tags causes silent metric drops
  • hot-shots buffers metrics and flushes at intervals — use client.close() on process exit to flush remaining metrics before shutdown
  • DogStatsD socket (UDS mode) requires Datadog Agent running locally with the dogstatsd_socket configured — UDP is the fallback for remote agents
  • Metric cardinality explosion from dynamic tags (user IDs, request IDs as tags) causes Datadog cost spikes — use sampling and controlled tag cardinality in agent metrics

Alternatives

Full Evaluation Report

Detailed scoring breakdown, competitive positioning, security analysis, and improvement recommendations for hot-shots.

$99

Scores are editorial opinions as of 2026-03-06.

5215
Packages Evaluated
26151
Need Evaluation
173
Need Re-evaluation
Community Powered