Vercel

Download a cache artifact

Downloads a cache artifact identified by its `hash` specified on the request path. The artifact is downloaded as an octet-stream (binary data). The artifact data is a gzip-compressed tarball containing the cached task outputs. The client will verify the content-length header against the response body size.

GET
/artifacts/{hash}

Authorization

bearerToken
AuthorizationBearer <token>

Bearer token authentication. Pass your authentication token in the Authorization header as Bearer <token>.

For self-hosted implementations, the token format and validation logic is up to the implementer. Common approaches include:

  • Static tokens for simple setups
  • JWT tokens for stateless authentication
  • OAuth2 tokens for integration with identity providers

In: header

Path Parameters

hash*string

The artifact hash. This is a content-addressable hash that uniquely identifies the cached artifact based on task inputs.

Match^[a-fA-F0-9]+$
Length1 <= length

Query Parameters

teamId?string

The team identifier to perform the request on behalf of. Used for multi-tenant cache implementations where artifacts are scoped to teams.

slug?string

The team slug to perform the request on behalf of. Alternative to teamId for identifying the team.

Header Parameters

x-artifact-client-ci?string

The continuous integration or delivery environment where this artifact operation is being performed. Examples: github-actions, circleci, jenkins.

Lengthlength <= 50
x-artifact-client-interactive?integer

Indicates whether the client is running in an interactive shell. 1 for interactive, 0 for non-interactive (CI/scripts).

Value in0 | 1

Response Body

application/octet-stream

application/json

application/json

application/json

application/json

curl -X GET "https://localhost:3000/artifacts/string"
"string"
{
  "code": "string",
  "message": "string"
}
{
  "code": "string",
  "message": "string"
}
{
  "code": "string",
  "message": "string"
}
{
  "code": "string",
  "message": "string"
}