MCP server that implements the Model Context Protocol to allow AI systems to interact with Vercel Operations
Server has 9 tools with basic descriptions but critical gaps in schema completeness, parameter documentation, and error handling. Most tools lack input parameter descriptions, only list_deployments and set_vercel_token have documented params. No output schemas are documented. Descriptions are present but generic (avg ~60 chars, below 194-char baseline). No error recovery guidance. Tool naming is acceptable (verb_noun pattern) but lacks specificity for distinguishing similar operations. Security concern: set_vercel_token accepts token as parameter, violating secret-injection pattern.
Check the status of the MCP server
Get detailed information about a specific Vercel project by ID
Get information about the authenticated Vercel user
List all aliases for a specific deployment
List recent deployments for a project
List environment variables for a Vercel project
List all domains associated with a Vercel project
No output schemas documented for any tool. LLMs cannot predict response structure, forcing them to guess at field names and types for downstream tool chaining.
set_vercel_token accepts API token as a parameter. This violates secret-injection pattern, tokens will be logged in traces and prompt history. Must use environment variable or secure vault injection instead.
7 of 9 tools have no parameter descriptions. LLMs cannot infer what project_id, deployment_id, or limit parameters mean without explicit documentation.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 47 | <=2025-11-25 | v2 |
List all Vercel projects in your account
Set the Vercel API token for this server session
No error handling guidance. Tools return generic error objects without actionable recovery steps. E.g., 'API request failed with status code 401' does not tell LLM whether to retry, ask user for new token, or abort.
Tool descriptions lack context on when to use each tool. E.g., list_projects vs get_project_info are not distinguished, LLM cannot decide which to call without trial and error.