@vercel/agent-eval-playground
Advanced tools
Web-based playground for browsing agent-eval experiment results
A web-based dashboard for browsing agent-eval experiment results. Ships as a standalone Next.js app
# Run from your eval project root (where results/ and evals/ live)
npx @vercel/agent-eval-playground
# With options
npx @vercel/agent-eval-playground --results-dir ./results --evals-dir ./evals --port 3001
| Flag | Default | Description |
|---|---|---|
--results-dir <dir> | ./results | Path to experiment results directory |
--evals-dir <dir> | ./evals | Path to eval fixtures directory |
--port, -p <port> | 3000 | HTTP server port |
--help, -h | Show help |
| Route | Description |
|---|---|
/ | Dashboard with stats, recent experiments, eval previews |
/experiments | Full experiment list |
/experiments/[name]/[timestamp] | Experiment detail with per-eval breakdown and run results |
/evals | All eval fixtures |
/evals/[name] | Eval detail with prompt, files, and content |
/compare | Side-by-side comparison of two experiment runs |
/transcript/[experiment]/[timestamp]/[evalName]/[run] | Full agent transcript viewer |
lib/data.ts reads fs directly)ComparePage dropdowns, TranscriptViewer collapsibles)force-dynamic on all pages — data can change between requestsThe CLI (bin.mjs) resolves the bundled next binary, sets RESULTS_DIR and EVALS_DIR as environment variables, and spawns next dev from the package directory. Server Components in lib/data.ts read directly from the filesystem.
Published alongside @vercel/agent-eval via changesets. Run npx changeset, select @vercel/agent-eval-playground, and merge to main. The release workflow handles the rest via OIDC.
FAQs
Web-based playground for browsing agent-eval experiment results
The npm package @vercel/agent-eval-playground receives a total of 102 weekly downloads. As such, @vercel/agent-eval-playground popularity was classified as not popular.
We found that @vercel/agent-eval-playground demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 373 open source maintainers collaborating on the project.
Did you know?
Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.
Security News
A compromised npm publish token was used to push a malicious postinstall script in cline@2.3.0, affecting the popular AI coding agent CLI with 90k weekly downloads.
Product
Socket is now scanning AI agent skills across multiple languages and ecosystems, detecting malicious behavior before developers install, starting with skills.sh's 60,000+ skills.
Product
Socket now supports PHP with full Composer and Packagist integration, enabling developers to search packages, generate SBOMs, and protect their PHP dependencies from supply chain threats.