go-skyvern

Unofficial Go client for the Skyvern API, generated with oapi-codegen from the published OpenAPI document. Package skyvern (github.com/psyb0t/go-skyvern) provides typed REST methods for tasks, agents, runs, browser sessions and profiles, credentials, scripts, schedules, tags, artifacts and audit exports. Use when Go code needs Skyvern browser automation or already imports go-skyvern.

Unofficial, typed Go client for the Skyvern API, in package skyvern at the module root.

For setup, operations and response handling, see references/setup.md.

Security & safety

• Read SKYVERN_API_KEY from the environment and send it in x-api-key with a request editor. Never hardcode, commit or log credentials or authenticated requests.
• Use https://api.skyvern.com for Skyvern Cloud. Use staging or a self-hosted deployment only when explicitly selected. Request bodies can contain browser instructions, URLs, credential references and sensitive data; send only what the requested task needs.
• Starting tasks, agents or scripts can interact with external websites and consume credits. Run, cancel, delete or change resources only when the user requested the action and only for resources owned by that task.

When to use

• Go code needs to start or inspect Skyvern automation runs.
• Go code needs Skyvern agents, browser sessions, profiles, schedules or artifacts.
• The project already imports github.com/psyb0t/go-skyvern.

When NOT to use

• The code is not Go. This module ships only a Go package.
• You need a client maintained by Skyvern. This client is unofficial.
• You want to change generated client code. See the repository's AGENTS.md.

Finding endpoints and types

The examples below cover a few calls. Every operation in the spec has a generated method, and the spec's descriptions are on them as Go doc comments. Look names and fields up instead of guessing. After go get github.com/psyb0t/go-skyvern, from inside your module:

bash

go doc github.com/psyb0t/go-skyvern ClientWithResponsesInterface  # every endpoint method
go doc github.com/psyb0t/go-skyvern RunTaskV1RunTasksPostParams   # one method, params type or model
go doc -all github.com/psyb0t/go-skyvern                          # the whole package

The same documentation is at https://pkg.go.dev/github.com/psyb0t/go-skyvern.

Method names come from the spec's operation IDs. <OperationId>WithResponse takes the context, then the path parameters in order, then *<OperationId>Params when the operation has query or header parameters, then the request body when it has one. The JSON200 field (or JSON201 and so on, per declared status) holds the decoded body; it is nil for any other status. Many operations declare x-api-key as an optional header parameter, so their params type has an XApiKey field; pass nil params and let the request editor set the header.

Usage

go

auth := skyvern.WithRequestEditorFn(
	func(_ context.Context, req *http.Request) error {
		req.Header.Set("x-api-key", os.Getenv("SKYVERN_API_KEY"))

		return nil
	},
)

client, err := skyvern.NewClientWithResponses("https://api.skyvern.com", auth)
if err != nil {
	return err
}

res, err := client.GetVersionV1VersionGetWithResponse(ctx)
if err != nil {
	return err
}

_ = res.JSON200

Imports are context, net/http, os and github.com/psyb0t/go-skyvern. The complete compiling example is example_test.go.

Check StatusCode() and the matching status field before reading a response. A nil error means the request and decoding succeeded; HTTP error statuses remain in the response. Supply an http.Client with a timeout through WithHTTPClient when needed. The generated client does not poll runs or retry rate limits automatically.