dsh-opencode-quota-router

A DeepSeek Harness (dsh) plugin that spends an OpenCode Go subscription before the user's DeepSeek account balance, per model.

OpenCode meters every model against its own monthly limit (5-hour = 20% of it, weekly = 50%, monthly = 100%), so an exhausted DeepSeek V4 Pro budget must not drag DeepSeek V4.1 Flash back to the DeepSeek account.

What it does

• Every routed request prefers the OpenCode Go route (opencode-go).
• When one model's allowance runs out, only that model falls back to the DeepSeek account; other models keep using the plan.
• While anything is falling back, a zero-token GET /inference/go/v1/usage probe runs every 30 minutes (or at the rolling-window reset) and switches back automatically when the allowance recovers.
• No session reload: both switches happen at step boundaries through the agent/request / agent/request-error waterfalls. Streaming requests and other agents are never interrupted.

Both listeners register with prepend: true, because a Cordis waterfall returns the outermost listener's value; an observer appended later would otherwise shadow the rewrite.

Routes

Primary (OpenCode Go)	Fallback (DeepSeek account)
opencode-go/deepseek-v4.1-flash	deepseek-account/deepseek-flash
opencode-go/deepseek-v4-pro	deepseek-account/deepseek-v4-pro

The bundle also declares the opencode-go catalog route (pruned to the models the live endpoint serves) and an opencode-free route for the free models (beam-free, iceman-contributor-free).

Install

dsh plugin --profile <profile> add dsh-opencode-quota-router

Then store the OpenCode key through the credentials service (the web Models page does this) under the reference name OPENCODE_API_KEY, or export it in the launching environment.

Configuration

The bundle patch carries defaults; override the row opencode-quota-router in your profile patch:

- id: opencode-quota-router
  config:
    usageUrl: https://opencode.ai/inference/go/v1/usage
    apiKeyRef: OPENCODE_API_KEY
    probeIntervalMs: 1800000
    modelQuarantineMs: 1800000
    routes:
      - primary: { provider: opencode-go, model: deepseek-v4.1-flash }
        fallback: { provider: deepseek-account, model: deepseek-flash }

Optional fields: minProbeDelayMs, unconfirmedDelayMs, modelQuarantineMs, statePath.

State

$DSH_HOME/opencode-quota-router.json:

{ "version": 2, "mode": "primary", "quarantine": {}, "usage": { }, "updatedAt": 0 }

• mode: "fallback" — plan-wide fallback (bad credential or unattributable failure).
• quarantine — one entry per exhausted model; only that model falls back until until.

Editing the file is picked up before the next step, so it doubles as a manual override:

{ "version": 2, "mode": "fallback", "until": 1793446953969, "quarantine": {} }

Test

node test.mjs

Exercises routing in both directions, per-model quarantine, quarantine expiry, plan-wide fallback, and the "transient rate limit is not a quarantine" rule against a fake Cordis context.

License

MIT