Agent debugging just moved into the apm you already run. sentry's agent tracing, ga since september…
agent debugging just moved into the apm you already run. sentry's agent tracing, ga since september 11, records every model call, tool run and sub-agent handoff with token cost as a span in your existing traces.
the agent is now just another service to profile.
Context
Sentry's changelog of 11 September 2026 says Agent Tracing is generally available and captures every step of an agent run, model calls, tool executions, sub-agent handoffs, token usage and cost, as spans inside existing Sentry traces, in the same hierarchy as HTTP, DB and queue spans. Adding conversation IDs lets you view the message exchange between agents and users, and new views show conversation titles, a cost and conversation overview chart, and timeline and transcript views. The docs say a model call span records model, input, output, token usage, cost and duration. Docs pages seen in snippets say head-based sampling applies to agent spans and the Model Cost widget shows estimated costs.
GA on 11 September and the span contents are first-party and supported. Every depends on setup: sampling is decided at the root span and libraries without automatic support need manual instrumentation per the snippet, and cost is described as estimated. The changelog frames the feature around failures with linked traces, logs and errors, and no measurement of faster debugging was seen. That the agent is now just another service to profile is the author's opinion.
Related work
- Earlier note on Sentry agent tracing ↗Same Sentry feature.
Watch next
- Sampling and cost calculation details from the docs.
Sources
- Agent Tracing is now GA (Sentry changelog, 11 Sep 2026)sentry.io
- Agent tracing (Sentry docs)docs.sentry.io
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 20 September 2026 at 19:56 IST. Sources are the papers and datasets the note draws on.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →