chore: fork — remove vendor CI, repoint release checks to Neuron Gitea

packages/opencode/test/session/compaction.test.ts

@@ -1816,19 +1816,6 @@ describe("SessionNs.getUsage", () => {
     expect(result.cost).toBe(3 + 1.5)
   })
 
-  test("uses authoritative Copilot billed cost when provided", () => {
-    const result = SessionNs.getUsage({
-      model: createModel({
-        context: 100_000,
-        output: 32_000,
-        cost: { input: 3, output: 15, cache: { read: 0.3, write: 0.3 } },
-      }),
-      usage: usage({ inputTokens: 11_774, outputTokens: 39, totalTokens: 11_813 }),
-      metadata: { copilot: { totalNanoAiu: 4_473_525_000 } },
-    })
-
-    expect(result.cost).toBe(0.04473525)
-  })
 
   test("uses matching context cost tier before over-200k fallback", () => {
     const model = createModel({

packages/opencode/test/session/llm.test.ts

@@ -504,56 +504,6 @@ describe("session.llm.ai-sdk adapter", () => {
     expect(result.tokens.cache.read).toBe(200)
   })
 
-  test("captures Copilot billed usage from raw Anthropic message deltas per step", async () => {
-    const events = await adapt([
-      uncheckedAdapterEvent({
-        type: "raw",
-        rawValue: {
-          type: "message_delta",
-          copilot_usage: { total_nano_aiu: 4_473_525_000 },
-        },
-      }),
-      {
-        type: "finish-step",
-        response: { id: "msg_test", timestamp: new Date(0), modelId: "claude-sonnet-4.6" },
-        finishReason: "stop",
-        rawFinishReason: "end_turn",
-        usage: {
-          inputTokens: 11_774,
-          outputTokens: 39,
-          totalTokens: 11_813,
-          inputTokenDetails: { noCacheTokens: 3, cacheReadTokens: 0, cacheWriteTokens: 11_771 },
-          outputTokenDetails: { textTokens: 39, reasoningTokens: undefined },
-        },
-        providerMetadata: { anthropic: { cacheCreationInputTokens: 11_771 } },
-      },
-      {
-        type: "finish-step",
-        response: { id: "msg_follow_up", timestamp: new Date(0), modelId: "claude-sonnet-4.6" },
-        finishReason: "stop",
-        rawFinishReason: "end_turn",
-        usage: {
-          inputTokens: 1,
-          outputTokens: 1,
-          totalTokens: 2,
-          inputTokenDetails: { noCacheTokens: 1, cacheReadTokens: 0, cacheWriteTokens: 0 },
-          outputTokenDetails: { textTokens: 1, reasoningTokens: undefined },
-        },
-        providerMetadata: { anthropic: {} },
-      },
-    ])
-
-    expect(events[0]).toMatchObject({
-      type: "step-finish",
-      providerMetadata: {
-        anthropic: { cacheCreationInputTokens: 11_771 },
-        copilot: { totalNanoAiu: 4_473_525_000 },
-      },
-    })
-    expect(events[1]).toMatchObject({ type: "step-finish", providerMetadata: { anthropic: {} } })
-    if (events[1].type !== "step-finish") throw new Error("expected step-finish")
-    expect(events[1].providerMetadata?.copilot).toBeUndefined()
-  })
 })
 
 type Capture = {

packages/opencode/test/session/system.test.ts

@@ -84,30 +84,23 @@ const it = testEffect(
 )
 
 describe("session.system", () => {
-  test("selects the Meta prompt for Muse Spark model IDs", () => {
-    for (const id of ["meta/muse-spark-preview", "muse-spark-1.1", "muse-spark-1.2"]) {
-      const prompt = SystemPrompt.provider({ api: { id } } as Provider.Model)[0]
-      expect(prompt).toContain("powered by Muse Spark,")
-      expect(prompt).toContain("using Meta Muse Spark.")
-      expect(prompt).not.toContain("{{MODEL_NAME}}")
+  test("uses the Neuron prompt for every model", () => {
+    const models = [
+      { providerID: "meta", api: { id: "muse-spark-preview" } },
+      { providerID: "moonshotai", api: { id: "k3" } },
+      { providerID: "anthropic", api: { id: "claude-sonnet-4-6" } },
+      { providerID: "openai", api: { id: "gpt-5.2" } },
+      { providerID: "google", api: { id: "gemini-3-pro" } },
+      { providerID: "mystery", api: { id: "unknown-model" } },
+    ]
+    for (const model of models) {
+      const prompt = SystemPrompt.provider(model as Provider.Model)
+      expect(prompt).toHaveLength(1)
+      expect(prompt[0]).toContain("You are Neuron")
+      expect(prompt[0]).not.toContain("{{MODEL_NAME}}")
     }
   })
 
-  test("selects the Meta prompt for Muse Glimmer model IDs", () => {
-    for (const id of ["meta/muse-glimmer", "meta/muse-glimmer-30b", "muse-glimmer-30b"]) {
-      const prompt = SystemPrompt.provider({ api: { id } } as Provider.Model)[0]
-      expect(prompt).toContain("powered by Muse Glimmer,")
-      expect(prompt).toContain("using Meta Muse Glimmer.")
-      expect(prompt).not.toContain("{{MODEL_NAME}}")
-    }
-  })
-
-  test("selects the Kimi prompt for official provider model IDs", () => {
-    for (const providerID of ["kimi-for-coding", "moonshotai", "moonshotai-cn"]) {
-      const prompt = SystemPrompt.provider({ providerID, api: { id: "k3" } } as Provider.Model)[0]
-      expect(prompt).toContain("# Prompt and Tool Use")
-    }
-  })
 
   it.effect("skills output is sorted by name and stable across calls", () =>
     Effect.gen(function* () {