/** * Reproducer for snapshot race condition with instant tool execution. * * When the mock LLM returns a tool call response instantly, the AI SDK * processes the tool call and executes the tool (e.g. apply_patch) before * the processor's start-step handler can capture a pre-tool snapshot. * Both the "before" and "after" snapshots end up with the same git tree * hash, so computeDiff returns empty and the session summary shows 0 files. * * This is a real bug: the snapshot system assumes it can capture state * before tools run by hooking into start-step, but the AI SDK executes * tools internally during multi-step processing before emitting events. */ import { expect } from "bun:test" import { Effect, Layer } from "effect" import { LayerNode } from "@opencode-ai/core/effect/layer-node" import fs from "fs/promises" import path from "path" import { Session } from "@/session/session" import { SessionPrompt } from "../../src/session/prompt" import { SessionSummary } from "../../src/session/summary" import { MessageV2 } from "../../src/session/message-v2" import { SessionV1 } from "@opencode-ai/core/v1/session" import { Database } from "@opencode-ai/core/database/database" import { SessionProjector } from "@opencode-ai/core/session/projector" import { provideTmpdirServer } from "../fixture/fixture" import { testEffect } from "../lib/effect" import { TestLLMServer } from "../lib/llm-server" import { LSP } from "@/lsp/lsp" import { MCP } from "../../src/mcp" import { CrossSpawnSpawner } from "@opencode-ai/core/cross-spawn-spawner" import { RuntimeFlags } from "@/effect/runtime-flags" const mcp = Layer.succeed( MCP.Service, MCP.Service.of({ status: () => Effect.succeed({}), clients: () => Effect.succeed({}), instructions: () => Effect.succeed([]), tools: () => Effect.succeed({}), prompts: () => Effect.succeed({}), resources: () => Effect.succeed({}), resourceTemplates: () => Effect.succeed({}), add: () => Effect.succeed({ status: { status: "disabled" as const } }), connect: () => Effect.void, disconnect: () => Effect.void, getPrompt: () => Effect.succeed(undefined), readResource: () => Effect.succeed(undefined), startAuth: () => Effect.die("unexpected MCP auth"), authenticate: () => Effect.die("unexpected MCP auth"), finishAuth: () => Effect.die("unexpected MCP auth"), removeAuth: () => Effect.void, supportsOAuth: () => Effect.succeed(false), hasStoredTokens: () => Effect.succeed(false), getAuthStatus: () => Effect.succeed("not_authenticated" as const), }), ) const lsp = Layer.succeed( LSP.Service, LSP.Service.of({ init: () => Effect.void, status: () => Effect.succeed([]), hasClients: () => Effect.succeed(false), touchFile: () => Effect.void, diagnostics: () => Effect.succeed({}), hover: () => Effect.succeed(undefined), definition: () => Effect.succeed([]), references: () => Effect.succeed([]), implementation: () => Effect.succeed([]), documentSymbol: () => Effect.succeed([]), workspaceSymbol: () => Effect.succeed([]), prepareCallHierarchy: () => Effect.succeed([]), incomingCalls: () => Effect.succeed([]), outgoingCalls: () => Effect.succeed([]), }), ) const root = LayerNode.group([ SessionPrompt.node, Session.node, SessionProjector.node, SessionSummary.node, Database.node, CrossSpawnSpawner.node, LayerNode.make({ service: TestLLMServer, layer: TestLLMServer.layer, deps: [] }), ]) const it = testEffect( LayerNode.compile(root, [ [MCP.node, mcp], [LSP.node, lsp], [RuntimeFlags.node, RuntimeFlags.layer({ experimentalEventSystem: true })], ]), ) const providerCfg = (url: string) => ({ provider: { test: { name: "Test", id: "test", env: [], npm: "@ai-sdk/openai-compatible", models: { "test-model": { id: "test-model", name: "Test Model", attachment: false, reasoning: false, temperature: false, tool_call: true, release_date: "2025-01-01", limit: { context: 100000, output: 10000 }, cost: { input: 0, output: 0 }, options: {}, }, }, options: { apiKey: "test-key", baseURL: url, }, }, }, }) it.live("tool execution produces non-empty session diff (snapshot race)", () => provideTmpdirServer( Effect.fnUntraced(function* ({ dir, llm }) { const prompt = yield* SessionPrompt.Service const sessions = yield* Session.Service const summary = yield* SessionSummary.Service const session = yield* sessions.create({ title: "snapshot race test", permission: [{ permission: "*", pattern: "*", action: "allow" }], }) // Use bash tool (always registered) to create a file const command = `echo 'snapshot race test content' > ${path.join(dir, "race-test.txt")}` yield* llm.toolMatch((hit) => JSON.stringify(hit.body).includes("create the file"), "bash", { command, }) yield* llm.textMatch((hit) => JSON.stringify(hit.body).includes("bash"), "done") // Seed user message yield* prompt.prompt({ sessionID: session.id, agent: "build", noReply: true, parts: [{ type: "text", text: "create the file" }], }) // Run the agent loop const result = yield* prompt.loop({ sessionID: session.id }) expect(result.info.role).toBe("assistant") // Verify the file was created const filePath = path.join(dir, "race-test.txt") const fileExists = yield* Effect.promise(() => fs .access(filePath) .then(() => true) .catch(() => false), ) expect(fileExists).toBe(true) // Verify the tool call completed (in the first assistant message) const allMsgs = yield* MessageV2.filterCompactedEffect(session.id) const user = allMsgs.find( (msg): msg is SessionV1.WithParts & { info: SessionV1.User } => msg.info.role === "user", ) const tool = allMsgs .flatMap((m) => m.parts) .find((p): p is SessionV1.ToolPart => p.type === "tool" && p.tool === "bash") expect(tool?.state.status).toBe("completed") if (!user) throw new Error("Expected user message") // Poll for the turn diff — summarize() is fire-and-forget. let diff: Array<{ file?: string }> = [] for (let i = 0; i < 50; i++) { diff = yield* summary.diff({ sessionID: session.id, messageID: user.info.id }) if (diff.length > 0) break yield* Effect.sleep("100 millis") } expect(diff.length).toBeGreaterThan(0) }), { git: true, config: providerCfg }, ), )