sentry-feature-setup
getsentry/sentry-for-ai
Route to the right Sentry feature skill—AI monitoring, OpenTelemetry, alerts, span streaming, or instrumentation guidance.
What is sentry-feature-setup?
A router skill that helps you identify and fetch the correct Sentry feature configuration skill based on your use case. Use this when you need to set up AI/LLM monitoring, OpenTelemetry pipelines, alerts, span streaming, or decide which observability signal to instrument.
- Routes requests to the correct feature skill (AI monitoring, OTel, alerts, span streaming, or instrumentation guide)
- Provides decision logic to identify which Sentry capability you need before diving into setup
- Explains how to fetch skills via curl from skills.sentry.dev
- Covers AI/LLM tracing, OpenTelemetry Collector integration, alert workflows, and span streaming across JavaScript, Python, and Dart/Flutter
How to install sentry-feature-setup
npx skills add https://github.com/getsentry/sentry-for-ai --skill sentry-feature-setupHow to use sentry-feature-setup
- 1.Read the decision tree in the 'Start Here' section to identify your use case
- 2.Ask the user which feature they want to configure if unclear
- 3.Use curl to fetch the appropriate skill: curl -sL https://skills.sentry.dev/{skill-path}/SKILL.md
- 4.Follow the instructions in the fetched skill for your specific feature
Use cases
- You need to monitor OpenAI, Anthropic, LangChain, or other AI SDK calls and don't know which skill to use
- You want to set up an OpenTelemetry Collector with Sentry exporter for multi-service telemetry routing
- You need to create alerts and notifications via Slack, PagerDuty, Discord, or email
- You're migrating from transaction-based to streamed span delivery in your JavaScript, Python, or Dart app
- You're unsure whether to instrument errors, traces, logs, or metrics for a given use case
- Engineers setting up Sentry for AI/LLM applications
- DevOps and platform teams configuring observability pipelines
- Teams implementing OpenTelemetry with Sentry
- Developers uncertain about which observability signal to use
sentry-feature-setup FAQ
Use the decision tree in the 'Start Here' section. If the user mentions AI/LLM monitoring, ask for `sentry-setup-ai-monitoring`. For OpenTelemetry, use `sentry-otel-exporter-setup`. For alerts, use `sentry-create-alert`. For span streaming, choose the language-specific skill. If unsure about signals, use `sentry-instrumentation-guide`.
No. Always append the exact path from the table to https://skills.sentry.dev/. Do not guess or shorten URLs.
Ask them first. Do not assume. Use the decision tree to guide them toward the right feature based on their description.
No. This is a router skill that helps you identify and fetch the correct feature skill. Each feature skill contains its own setup instructions.
JavaScript, Python, and Dart/Flutter each have their own language-specific span streaming skill.
Full instructions (SKILL.md)
Source of truth, from getsentry/sentry-for-ai.
name: sentry-feature-setup description: Configure specific Sentry features beyond basic SDK setup. Use when asked to monitor AI/LLM calls, set up OpenTelemetry pipelines, create alerts and notifications, or enable span streaming. license: Apache-2.0 role: router
Sentry Feature Setup
Configure specific Sentry capabilities beyond basic SDK setup — AI monitoring, OpenTelemetry pipelines, alerts, and deciding which signal to instrument where. This page helps you find the right feature skill for your task.
How to Fetch Skills
Use curl to download skills — they are 10–20 KB files that fetch tools often summarize, losing critical details.
curl -sL https://skills.sentry.dev/sentry-setup-ai-monitoring/SKILL.md
Append the path from the Path column in the table below to https://skills.sentry.dev/. Do not guess or shorten URLs.
Start Here — Read This Before Doing Anything
Do not skip this section. Do not assume which feature the user needs. Ask first.
- If the user mentions AI monitoring, LLM tracing, conversations, or instrumenting an AI SDK (OpenAI, Anthropic, LangChain, Vercel AI, Google GenAI, Pydantic AI) →
sentry-setup-ai-monitoring - If the user mentions OpenTelemetry, OTel Collector, or multi-service telemetry routing →
sentry-otel-exporter-setup - If the user mentions alerts, notifications, on-call, Slack/PagerDuty/Discord integration, or workflow rules →
sentry-create-alert - If the user mentions span streaming, traceLifecycle, trace_lifecycle, spanStreamingIntegration, or switching from transactions to streamed spans →
sentry-span-streaming-js(JavaScript),sentry-span-streaming-python(Python), orsentry-span-streaming-dart(Dart/Flutter) - If the user is unsure which signal to use — log vs span vs metric, "what to instrument where", or how to choose between errors, traces, logs, and metrics →
sentry-instrumentation-guide
When unclear, ask the user which feature they want to configure. Do not guess.
Feature Skills
| Feature | Skill | Path |
|---|---|---|
| AI/LLM monitoring and conversations — instrument OpenAI, Anthropic, LangChain, Vercel AI, Google GenAI, Pydantic AI | sentry-setup-ai-monitoring | sentry-setup-ai-monitoring/SKILL.md |
| OpenTelemetry Collector with Sentry Exporter — multi-project routing, automatic project creation | sentry-otel-exporter-setup | sentry-otel-exporter-setup/SKILL.md |
| Alerts via workflow engine API — email, Slack, PagerDuty, Discord | sentry-create-alert | sentry-create-alert/SKILL.md |
| Span streaming (JavaScript) — migrate from transaction-based to streamed span delivery | sentry-span-streaming-js | sentry-span-streaming-js/SKILL.md |
| Span streaming (Python) — migrate from transaction-based to streamed span delivery | sentry-span-streaming-python | sentry-span-streaming-python/SKILL.md |
| Span streaming (Dart/Flutter) — migrate from transaction-based to streamed span delivery | sentry-span-streaming-dart | sentry-span-streaming-dart/SKILL.md |
| Instrumentation guide — decide which signal to reach for (error vs span vs log vs metric), "what to instrument where" | sentry-instrumentation-guide | sentry-instrumentation-guide/SKILL.md |
Each skill contains its own detection logic, prerequisites, and step-by-step instructions. Trust the skill — read it carefully and follow it. Do not improvise or take shortcuts.
Looking for SDK setup or debugging workflows instead? See the full Skill Tree.
Related skills
More from getsentry/sentry-for-ai and the wider catalog.
sentry-fix-issues
Find and fix production issues from Sentry by analyzing stack traces, breadcrumbs, and traces to identify root causes.
sentry-react-native-sdk
Full Sentry SDK setup for React Native and Expo with error monitoring, tracing, profiling, and session replay.
sentry-workflow
Route to the right Sentry workflow for fixing production issues, reviewing code, or upgrading SDKs.
sentry-nextjs-sdk
Full Sentry SDK setup for Next.js with error monitoring, tracing, session replay, and AI monitoring.
sentry-sdk-setup
Detect your platform and route to the right Sentry SDK skill for error monitoring setup.
sentry-create-alert
Create Sentry alerts via workflow engine API with email, Slack, PagerDuty, Discord, and other notifications.