# Automated agent triage with Agent Tracing and Claude Routines

DevFeed: [Automated agent triage with Agent Tracing and Claude Routines](<https://devfeed.tech/articles/automated-agent-triage-with-agent-tracing-and-claude-routines-24091.md>)

Original publisher: [Read original article](<https://blog.sentry.io/claude-routines-agent-triage/>)

Author: Trevor Elkins

Published: 2026-08-13T09:00:00Z

Content type: article

Language: en

Sources: [Sentry](<https://devfeed.tech/sources/sentry-blog.md>)

Topics: [agent observability](<https://devfeed.tech/topics/agent-observability.md>), [tracing](<https://devfeed.tech/topics/tracing.md>), [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [Claude](<https://devfeed.tech/topics/claude.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-loops](<https://devfeed.tech/tags/agent-loops.md>), [agent-tracing](<https://devfeed.tech/tags/agent-tracing.md>), [aggregate](<https://devfeed.tech/tags/aggregate.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [automated](<https://devfeed.tech/tags/automated.md>), [automatically](<https://devfeed.tech/tags/automatically.md>), [behavior](<https://devfeed.tech/tags/behavior.md>), [bugs](<https://devfeed.tech/tags/bugs.md>), [claude](<https://devfeed.tech/tags/claude.md>), [conversations](<https://devfeed.tech/tags/conversations.md>), [file](<https://devfeed.tech/tags/file.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [routine](<https://devfeed.tech/tags/routine.md>), [routines](<https://devfeed.tech/tags/routines.md>), [sentry](<https://devfeed.tech/tags/sentry.md>), [tracing](<https://devfeed.tech/tags/tracing.md>), [triage](<https://devfeed.tech/tags/triage.md>)

## AI overview

Sentry's Claude Routine uses the Sentry MCP to query Agent Tracing data from the previous night and triage about 800 Seer conversations each morning. It gathers conversation totals and tool error rates, samples failed-tool spans and full conversations with and without tool errors, checks whether Seer's verdict matches its reasoning, and consults the codebase for context. It then reports new findings and searches Linear to avoid filing duplicate tickets. The workflow is intended to surface tool failures, hallucinated outputs, latency spikes, agent loops and cost anomalies, while also checking that apparently good runs are sound. The excerpt says this replaces a manual process involving limited queries and hand-written parsing of raw trace JSON; it does not provide measured outcomes or details beyond the described workflow.

## Source excerpt

How Sentry uses a Claude Routine and the Sentry MCP to automatically triage 800 AI agent conversations overnight and file bugs.