Vercel launched Is Agentic, a free web‑based scoring service that evaluates how easily an AI agent can navigate a public site. The tool is powered by Ora’s agent‑experience research, which supplies a checklist of over one hundred verification items. Users can obtain a score by entering a URL in the browser or invoking the provided CLI; no API key, subscription, or usage fees are required, and the underlying infrastructure—interface, report storage, and aggregation—is operated by Vercel.
The scoring reflects Ora’s four‑layer framework: discovery (20 pts, 15 checks), access (30 pts, 41 checks), usability (40 pts, 56 checks), and payments (10 pts, 6 checks), totaling 118 checks that underlie Vercel’s “100+” claim. Ora converts points to letter grades (A+ ≥ 95, A 86‑94, B 70‑85, C 48‑69, D 28‑47, F < 28). Vercel then groups checks into essential (80 pts), recommended (20 pts), and emerging signals (up to 5 bonus pts), excludes not‑applicable items, awards partial credit proportionally, and averages duplicate check IDs across MCP surfaces.
- Free to use; no API key, subscription, or per‑report fees.
- Accessible via web UI, read‑only API, CLI, or MCP server.
- Scores based on Ora’s 118‑check agent‑experience checklist.
- Letter‑grade output (A+ to F) with Vercel’s essential/recommended/emerging weighting.
- Enables CI regression gates and cross‑benchmarking of public sites.
Why this matters
By providing a zero‑cost, transparent audit that mirrors actual agent behavior, Is Agentic lets teams embed agent‑readiness checks into CI pipelines and compare public sites on a common scale. While the tool lowers the friction for improving discoverability, its real‑world impact will depend on adoption rates and subsequent remediation of flagged issues.
Share this article
Found this insightful? Share it with your community on Reddit, X, or copy the link.