AI Agent Hub
Back to plugins
🖥️

dsh-opencode-session-id

Client Updated 2026.08.24

Run the following command in DeepSeek Harness:

dsh plugin install gausszhou/dsh-opencode-session-id

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

Install it in DeepSeek Harness by running dsh plugin install gausszhou/dsh-opencode-session-id; the source is available at https://github.com/gausszhou/dsh-opencode-session-id

About this plugin

When routing requests through an opencode gateway (opencode, opencode-go, etc.) via DeepSeek Harness, outgoing HTTP calls lack the session-id headers that the native opencode client normally sends — breaking per-session tracing and concurrency isolation on the gateway side. dsh-opencode-session-id closes that gap: it wraps the fetch layer for matched opencode endpoints and injects x-opencode-session, x-session-affinity, and related headers so every outbound request carries a session identifier just like the opencode client itself.

It works out of the box with zero configuration. The uuid portion of the session id (session- prefix excluded) is hashed via SHA-256 into a short, irreversible alphanumeric nanoid(8) token (e.g. 0RpJJnxJ), so the backend never sees the raw id. The plugin only appends request headers — body, URL, and method pass through unchanged. Concurrent sessions stay isolated, and non-opencode endpoints are untouched. For self-hosted gateways you can extend matching via baseURLs or add fingerprint headers like x-opencode-client through extraHeaders.

Built for anyone routing opencode models through dsh who needs session-level association at the gateway, as well as developers running a custom opencode proxy that relies on session headers for routing or rate-limiting. Install, restart dsh web, and you are done.

Use Cases

  • Injecting session-id headers into outbound opencode gateway requests via dsh
  • Self-hosted opencode proxy relying on session headers for routing or rate-limiting
  • Keeping per-session request isolation across concurrent sessions at the gateway

Best For

  • Users routing opencode models through DeepSeek Harness
  • Developers building custom opencode proxies
  • LLM infrastructure teams needing gateway-side session association