# Trace and debug eve agent sessions with Vercel Observability

DevFeed: [Trace and debug eve agent sessions with Vercel Observability](<https://devfeed.tech/articles/trace-and-debug-eve-agent-sessions-with-vercel-observability-920.md>)

Original publisher: [Read original article](<https://vercel.com/changelog/eve-agent-observability>)

Author: John Phamous

Published: 2026-06-26T00:00:00Z

Content type: release

Language: en

Sources: [Vercel News](<https://devfeed.tech/sources/vercel-news.md>)

Topics: [agent observability](<https://devfeed.tech/topics/agent-observability.md>), [observability](<https://devfeed.tech/topics/observability.md>), [Vercel](<https://devfeed.tech/topics/vercel.md>), [OpenTelemetry](<https://devfeed.tech/topics/opentelemetry.md>), [vercel ai sdk](<https://devfeed.tech/topics/vercel-ai-sdk.md>), [debugging](<https://devfeed.tech/topics/debugging.md>), [Open Source](<https://devfeed.tech/topics/open-source.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [debug](<https://devfeed.tech/tags/debug.md>), [instrumentation](<https://devfeed.tech/tags/instrumentation.md>), [json](<https://devfeed.tech/tags/json.md>), [logs](<https://devfeed.tech/tags/logs.md>), [observability](<https://devfeed.tech/tags/observability.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [opentelemetry](<https://devfeed.tech/tags/opentelemetry.md>), [sdk](<https://devfeed.tech/tags/sdk.md>), [token](<https://devfeed.tech/tags/token.md>), [vercel](<https://devfeed.tech/tags/vercel.md>)

## AI overview

Vercel Observability now provides Agent Runs for eve projects, letting teams inspect agent sessions, model calls, tool calls, errors, token usage, and execution details. Developer and Business views support technical debugging and non-technical auditing, while encrypted storage, configurable retention, and AI SDK span export are supported.

## Source excerpt

You can now view Agent Runs in your Vercel dashboard for eve, the open-source agent framework. The Agent Runs tab appears automatically for every eve project, surfacing trigger, duration, and token usage for each session at a glance. Drill into any run to inspect every turn, model call, and tool call in the conversation. Runtime errors that previously vanished into function logs now correlate to the failing step. Two views read the same data: Developer mode shows raw tool names, input and output JSON, and per-step token counts for engineers debugging a failure. Business mode humanizes tool names, hides JSON, and generates a plain-English summary so a non-technical reviewer can audit what an agent did. Run data is encrypted by default. Retention scales with plan: Hobby: 12 hours Pro: 1 day Enterprise: 3 days Pro and Enterprise teams can extend retention to 30 days with Observability Plus. Teams using a custom OpenTelemetry backend can still export AI SDK spans from agent/instrumentation.ts to any destination, such as Braintrust or Datadog. Open Agent Runs in your project and read the documentation to get started. Read more