Skip to main content

Agent Runs now available in the Vercel MCP and CLI

By Steven Van ·

New Vercel MCP tools and CLI commands let developers and agents inspect eve agent traces, reasoning, and token usage.

The Vercel MCP server and CLI can now inspect Agent Runs, the observability layer for agents built with eve, Vercel's open-source agent framework. Traces from eve agents deployed to Vercel are ingested automatically, and the new tools let an agent (or a developer) find projects with runs, list recent runs, and pull full traces, including reasoning, tool calls, and token usage.

  • list_agent_run_projects: find projects in a team with Agent Runs activity
  • list_agent_runs: list recent runs for a project
  • get_agent_run: inspect a run's metadata, lifecycle events, usage, and subagent data
  • get_agent_run_trace: retrieve a run's full trace, including turns, messages, reasoning, tool calls, and tool input/output

The same data is available from the CLI via vercel agent-runs projects, list, inspect, and trace. Every subcommand supports --json, and traces render as markdown when piped, so coding agents without MCP access can call the CLI directly to debug their own runs. Details are in the announcement.

Vercel
Vercel
The platform for frontend developers — deploy, preview, and scale web apps and AI agents with zero config.
View Vercel →

Read the original announcement →

Read Agent Runs now available in the Vercel MCP and CLI on Creators Toolbox