AI Agent Hub
Back to plugins
🖥️

dsh-asc

Client Updated 2026.08.29

Run the following command in DeepSeek Harness:

dsh plugin install Tangweiwei227/dsh-asc

Paste the following prompt into your AI chat to install this plugin:

Run dsh plugin install Tangweiwei227/dsh-asc to install this plugin in DeepSeek Harness; the full source is available at https://github.com/Tangweiwei227/dsh-asc .

About this plugin

Managing an App Store Connect account usually means running commands like asc apps list, asc build upload, or asc metadata pull in a terminal, while having a model build raw shell strings inside DSH introduces injection risk. dsh-asc wraps the open-source App Store Connect CLI as a structured agent tool: the model passes a string array (e.g. ["apps","list"]) and the plugin invokes the binary via execFile with no shell involved, eliminating command injection by design.

Behind that single tool sit 76 top-level command groups covering over 1,200 App Store Connect API endpoints: TestFlight build uploads, App Store releases and rollbacks, Apple Ads, code signing, localized metadata and keywords, subscription management, analytics, Xcode Cloud, and alternative distribution. In non-interactive mode asc emits JSON automatically; the plugin passes the output through verbatim as { ok, exitCode, stdout, stderr } so the model can parse it precisely. Destructive operations still require asc's own --confirm flag, adding an extra safety rail.

If you already manage iOS or macOS app releases, metadata, or ads within a DSH workflow and would rather not maintain a pile of shell scripts, dsh-asc lets you drive everyday tasks with plain natural language while keeping the call chain safe and auditable.

Use Cases

  • Query all apps and build statuses under an account in natural language
  • Bulk pull localized description and keyword metadata
  • Upload a new build to TestFlight and track release progress

Best For

  • iOS/macOS developers who frequently interact with App Store Connect
  • Teams managing app releases and metadata within DSH workflows
  • AI agent users who want to avoid shell injection risks