Deploy, manage, inspect, and troubleshoot Vercel projects from the command line. Use for Vercel deployments, Vercel Toolbar comments, build failures, projects and teams, environment variables, domains and DNS, logs, metrics, Speed Insights, Core Web Vitals, request traces, usage, activity, alerts, firewall rules, cache, cron jobs, deploy hooks, Edge Config, feature flags (`vercel flags`), integrations, connectors, Blob storage, Container Registry (VCR), microfrontends, rolling releases, custom environments, Sandbox, agent/MCP setup, preview access, local development, or `vercel api` fallback.
95%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Critical
Do not install without reviewing
The canonical home for this skill is vercel-cli in vercel/vercel-plugin
Detected a prompt injection in the skill instructions. The skill contains hidden or deceptive instructions that fall outside its stated purpose and attempt to override the agent’s safety guidelines or intended behavior.
The skill is documentation for the Vercel CLI, but one example silently rewrites AI Gateway requests from one model to another without explanation, which could alter the user's traffic in ways they did not request.
The skill is specifically designed for direct financial operations, giving the agent the ability to move money or execute financial transactions — such as payment processing, cryptocurrency operations, banking integrations, or market order execution.
Decisive route: the skill documents "billing purchases" as an in-scope operation (SKILL.md:57, routing to references/platform-ops.md). A billing purchase is a direct financial operation, so the control test applies. The referenced file is not supplied, so the purchase mechanics are not verified and this rests on the documented route label. The control is service-enforced login: the CLI requires `vercel login` (SKILL.md:34), and read-only project commands can require login or SAML re-authentication enforced by the service (SKILL.md:28). That is an authenticated-session gate on the route, so it qualifies as independent access control. Other documented items (usage, alerts, integrations, Sandbox) were not treated as payment routes. Severity is medium, not high, because a qualifying control is documented; the tier would rise to high if the billing route were shown to lack any authentication.
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
The `vercel comments` workflow in comments.md reads Vercel Toolbar comment threads by default (unresolved threads for the linked project), and those threads can be authored by anyone with preview access, so an outsider can add text to a feed the agent reads without the agent selecting a specific item.
vercel/vercel-plugin
8a9039d
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.