fix: add null guards for output.output in tool.execute.after hooks

Three hooks crashed with TypeError when MCP tools returned results where
output.output is undefined. Added type guards to all affected hooks:

- comment-checker/hook.ts: guard before toLowerCase()
- edit-error-recovery/hook.ts: guard before toLowerCase()
- task-resume-info/hook.ts: guard before startsWith()/includes()/trimEnd()
- Added test for undefined output.output in edit-error-recovery

Fixes #1746

.github/workflows/cla.yml

@@ -25,7 +25,7 @@ jobs:
           path-to-signatures: 'signatures/cla.json'
           path-to-document: 'https://github.com/code-yeongyu/oh-my-opencode/blob/master/CLA.md'
           branch: 'dev'
-          allowlist: code-yeongyu,bot*,dependabot*,github-actions*,*[bot],sisyphus-dev-ai
+          allowlist: code-yeongyu,bot*,dependabot*,github-actions*,*[bot],sisyphus-dev-ai,web-flow
           custom-notsigned-prcomment: |
             Thank you for your contribution! Before we can merge this PR, we need you to sign our [Contributor License Agreement (CLA)](https://github.com/code-yeongyu/oh-my-opencode/blob/master/CLA.md).
             

bun.lock

@@ -28,13 +28,13 @@
         "typescript": "^5.7.3",
       },
       "optionalDependencies": {
-        "oh-my-opencode-darwin-arm64": "3.3.1",
-        "oh-my-opencode-darwin-x64": "3.3.1",
-        "oh-my-opencode-linux-arm64": "3.3.1",
-        "oh-my-opencode-linux-arm64-musl": "3.3.1",
-        "oh-my-opencode-linux-x64": "3.3.1",
-        "oh-my-opencode-linux-x64-musl": "3.3.1",
-        "oh-my-opencode-windows-x64": "3.3.1",
+        "oh-my-opencode-darwin-arm64": "3.5.1",
+        "oh-my-opencode-darwin-x64": "3.5.1",
+        "oh-my-opencode-linux-arm64": "3.5.1",
+        "oh-my-opencode-linux-arm64-musl": "3.5.1",
+        "oh-my-opencode-linux-x64": "3.5.1",
+        "oh-my-opencode-linux-x64-musl": "3.5.1",
+        "oh-my-opencode-windows-x64": "3.5.1",
       },
     },
   },
@@ -226,19 +226,19 @@
 
     "object-inspect": ["object-inspect@1.13.4", "", {}, "sha512-W67iLl4J2EXEGTbfeHCffrjDfitvLANg0UlX3wFUUSTx92KXRFegMHUVgSqE+wvhAbi4WqjGg9czysTV2Epbew=="],
 
-    "oh-my-opencode-darwin-arm64": ["oh-my-opencode-darwin-arm64@3.3.1", "", { "os": "darwin", "cpu": "arm64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-R+o42Km6bsIaW6D3I8uu2HCF3BjIWqa/fg38W5y4hJEOw4mL0Q7uV4R+0vtrXRHo9crXTK9ag0fqVQUm+Y6iAQ=="],
+    "oh-my-opencode-darwin-arm64": ["oh-my-opencode-darwin-arm64@3.5.1", "", { "os": "darwin", "cpu": "arm64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-oH+c/+Z/ULIK+8T1jQFpzISHsvQPyYJfA6bceiD9sgFy1OY1NjRh4a3sFk8cXy6uRVKpivWDFOfbVTcZ2kbKWA=="],
 
-    "oh-my-opencode-darwin-x64": ["oh-my-opencode-darwin-x64@3.3.1", "", { "os": "darwin", "cpu": "x64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-7VTbpR1vH3OEkoJxBKtYuxFPX8M3IbJKoeHWME9iK6FpT11W1ASsjyuhvzB1jcxSeqF8ddMnjitlG5ub6h5EVw=="],
+    "oh-my-opencode-darwin-x64": ["oh-my-opencode-darwin-x64@3.5.1", "", { "os": "darwin", "cpu": "x64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-wnBYQ9BZBLbzgSNIJZOIJS03zf+b4trAQeYmG+yCLn8y7FWXqw1KmjJ88/bbMXTuZ4RSMKWpXb1Afgdsred+DQ=="],
 
-    "oh-my-opencode-linux-arm64": ["oh-my-opencode-linux-arm64@3.3.1", "", { "os": "linux", "cpu": "arm64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-BZ/r/CFlvbOxkdZZrRoT16xFOjibRZHuwQnaE4f0JvOzgK6/HWp3zJI1+2/aX/oK5GA6lZxNWRrJC/SKUi8LEg=="],
+    "oh-my-opencode-linux-arm64": ["oh-my-opencode-linux-arm64@3.5.1", "", { "os": "linux", "cpu": "arm64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-19KNJex1LeU/S14IsJbumOvZa9O6F7X4BLIY7MfjtHtTk0dRFL+tbbXmlafecBMigEKlLdJ+HTW3TnQgp7Ih8A=="],
 
-    "oh-my-opencode-linux-arm64-musl": ["oh-my-opencode-linux-arm64-musl@3.3.1", "", { "os": "linux", "cpu": "arm64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-U90Wruf21h+CJbtcrS7MeTAc/5VOF6RI+5jr7qj/cCxjXNJtjhyJdz/maehArjtgf304+lYCM/Mh1i+G2D3YFQ=="],
+    "oh-my-opencode-linux-arm64-musl": ["oh-my-opencode-linux-arm64-musl@3.5.1", "", { "os": "linux", "cpu": "arm64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-mCCnym3nBTJP+xzK+AS4YPFQiT2sZWmjhOhOy7PjNY6Is4jkfT1C2e9ZrIU/2VoVLV6V5q7hQGh1jgleU+FxwQ=="],
 
-    "oh-my-opencode-linux-x64": ["oh-my-opencode-linux-x64@3.3.1", "", { "os": "linux", "cpu": "x64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-sYzohSNdwsAhivbXcbhPdF1qqQi2CCI7FSgbmvvfBOMyZ8HAgqOFqYW2r3GPdmtywzkjOTvCzTG56FZwEjx15w=="],
+    "oh-my-opencode-linux-x64": ["oh-my-opencode-linux-x64@3.5.1", "", { "os": "linux", "cpu": "x64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-sDYt4adNuwb+p1RzHb7IR9zvbAnYYgZofjPvceirBorffp63f+aypYFxjFpfmbT87o/Eb/Hgzm4sHliJtd1UmQ=="],
 
-    "oh-my-opencode-linux-x64-musl": ["oh-my-opencode-linux-x64-musl@3.3.1", "", { "os": "linux", "cpu": "x64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-aG5pZ4eWS0YSGUicOnjMkUPrIqQV4poYF+d9SIvrfvlaMcK6WlQn7jXzgNCwJsfGn5lyhSmjshZBEU+v79Ua3w=="],
+    "oh-my-opencode-linux-x64-musl": ["oh-my-opencode-linux-x64-musl@3.5.1", "", { "os": "linux", "cpu": "x64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-tz/0QSS5AKIiKj6cMom5VQSnEYpMIP/SRTaP5WYNOYhnUkXMwXEncQ7FIcj2vovMCXuqA9a8ujVY0zTs7TeALw=="],
 
-    "oh-my-opencode-windows-x64": ["oh-my-opencode-windows-x64@3.3.1", "", { "os": "win32", "cpu": "x64", "bin": { "oh-my-opencode": "bin/oh-my-opencode.exe" } }, "sha512-FGH7cnzBqNwjSkzCDglMsVttaq+MsykAxa7ehaFK+0dnBZArvllS3W13a3dGaANHMZzfK0vz8hNDUdVi7Z63cA=="],
+    "oh-my-opencode-windows-x64": ["oh-my-opencode-windows-x64@3.5.1", "", { "os": "win32", "cpu": "x64", "bin": { "oh-my-opencode": "bin/oh-my-opencode.exe" } }, "sha512-zfpRS6HIkSwE8btajJzSYxhqsE5kDkop896/XGS3LLIAAZt0RtCmT3C1plxVfI9oAABfgcaiveCxJ5f9AlKPcQ=="],
 
     "on-finished": ["on-finished@2.4.1", "", { "dependencies": { "ee-first": "1.1.1" } }, "sha512-oVlzkg3ENAhCk2zdv7IJwd/QUD4z2RxRwpkcGY8psCVcCYZNq4wYnVWALHM+brtuJjePWiYF/ClmuDr8Ch5+kg=="],
 

package.json

@@ -1,6 +1,6 @@
 {
   "name": "oh-my-opencode",
-  "version": "3.5.0",
+  "version": "3.5.1",
   "description": "The Best AI Agent Harness - Batteries-Included OpenCode Plugin with Multi-Model Orchestration, Parallel Background Agents, and Crafted LSP/AST Tools",
   "main": "dist/index.js",
   "types": "dist/index.d.ts",
@@ -74,13 +74,13 @@
     "typescript": "^5.7.3"
   },
   "optionalDependencies": {
-    "oh-my-opencode-darwin-arm64": "3.5.0",
-    "oh-my-opencode-darwin-x64": "3.5.0",
-    "oh-my-opencode-linux-arm64": "3.5.0",
-    "oh-my-opencode-linux-arm64-musl": "3.5.0",
-    "oh-my-opencode-linux-x64": "3.5.0",
-    "oh-my-opencode-linux-x64-musl": "3.5.0",
-    "oh-my-opencode-windows-x64": "3.5.0"
+    "oh-my-opencode-darwin-arm64": "3.5.1",
+    "oh-my-opencode-darwin-x64": "3.5.1",
+    "oh-my-opencode-linux-arm64": "3.5.1",
+    "oh-my-opencode-linux-arm64-musl": "3.5.1",
+    "oh-my-opencode-linux-x64": "3.5.1",
+    "oh-my-opencode-linux-x64-musl": "3.5.1",
+    "oh-my-opencode-windows-x64": "3.5.1"
   },
   "trustedDependencies": [
     "@ast-grep/cli",

packages/darwin-arm64/package.json

@@ -1,6 +1,6 @@
 {
   "name": "oh-my-opencode-darwin-arm64",
-  "version": "3.5.0",
+  "version": "3.5.1",
   "description": "Platform-specific binary for oh-my-opencode (darwin-arm64)",
   "license": "MIT",
   "repository": {

packages/darwin-x64/package.json

@@ -1,6 +1,6 @@
 {
   "name": "oh-my-opencode-darwin-x64",
-  "version": "3.5.0",
+  "version": "3.5.1",
   "description": "Platform-specific binary for oh-my-opencode (darwin-x64)",
   "license": "MIT",
   "repository": {

packages/linux-arm64-musl/package.json

@@ -1,6 +1,6 @@
 {
   "name": "oh-my-opencode-linux-arm64-musl",
-  "version": "3.5.0",
+  "version": "3.5.1",
   "description": "Platform-specific binary for oh-my-opencode (linux-arm64-musl)",
   "license": "MIT",
   "repository": {

packages/linux-arm64/package.json

@@ -1,6 +1,6 @@
 {
   "name": "oh-my-opencode-linux-arm64",
-  "version": "3.5.0",
+  "version": "3.5.1",
   "description": "Platform-specific binary for oh-my-opencode (linux-arm64)",
   "license": "MIT",
   "repository": {

packages/linux-x64-musl/package.json

@@ -1,6 +1,6 @@
 {
   "name": "oh-my-opencode-linux-x64-musl",
-  "version": "3.5.0",
+  "version": "3.5.1",
   "description": "Platform-specific binary for oh-my-opencode (linux-x64-musl)",
   "license": "MIT",
   "repository": {

packages/linux-x64/package.json

@@ -1,6 +1,6 @@
 {
   "name": "oh-my-opencode-linux-x64",
-  "version": "3.5.0",
+  "version": "3.5.1",
   "description": "Platform-specific binary for oh-my-opencode (linux-x64)",
   "license": "MIT",
   "repository": {

packages/windows-x64/package.json

@@ -1,6 +1,6 @@
 {
   "name": "oh-my-opencode-windows-x64",
-  "version": "3.5.0",
+  "version": "3.5.1",
   "description": "Platform-specific binary for oh-my-opencode (windows-x64)",
   "license": "MIT",
   "repository": {

signatures/cla.json

@@ -1319,6 +1319,38 @@
       "created_at": "2026-02-10T15:32:31Z",
       "repoId": 1108837393,
       "pullRequestNo": 1723
+    },
+    {
+      "name": "sjawhar",
+      "id": 5074378,
+      "comment_id": 3879746658,
+      "created_at": "2026-02-10T17:43:47Z",
+      "repoId": 1108837393,
+      "pullRequestNo": 1727
+    },
+    {
+      "name": "marlon-costa-dc",
+      "id": 128386606,
+      "comment_id": 3879827362,
+      "created_at": "2026-02-10T17:59:06Z",
+      "repoId": 1108837393,
+      "pullRequestNo": 1726
+    },
+    {
+      "name": "marlon-costa-dc",
+      "id": 128386606,
+      "comment_id": 3879847814,
+      "created_at": "2026-02-10T18:03:41Z",
+      "repoId": 1108837393,
+      "pullRequestNo": 1726
+    },
+    {
+      "name": "danpung2",
+      "id": 75434746,
+      "comment_id": 3881834946,
+      "created_at": "2026-02-11T02:52:34Z",
+      "repoId": 1108837393,
+      "pullRequestNo": 1741
     }
   ]
 }

src/features/boulder-state/storage.test.ts

@@ -43,6 +43,78 @@ describe("boulder-state", () => {
       expect(result).toBeNull()
     })
 
+    test("should return null for JSON null value", () => {
+      //#given - boulder.json containing null
+      const boulderFile = join(SISYPHUS_DIR, "boulder.json")
+      writeFileSync(boulderFile, "null")
+
+      //#when
+      const result = readBoulderState(TEST_DIR)
+
+      //#then
+      expect(result).toBeNull()
+    })
+
+    test("should return null for JSON primitive value", () => {
+      //#given - boulder.json containing a string
+      const boulderFile = join(SISYPHUS_DIR, "boulder.json")
+      writeFileSync(boulderFile, '"just a string"')
+
+      //#when
+      const result = readBoulderState(TEST_DIR)
+
+      //#then
+      expect(result).toBeNull()
+    })
+
+    test("should default session_ids to [] when missing from JSON", () => {
+      //#given - boulder.json without session_ids field
+      const boulderFile = join(SISYPHUS_DIR, "boulder.json")
+      writeFileSync(boulderFile, JSON.stringify({
+        active_plan: "/path/to/plan.md",
+        started_at: "2026-01-01T00:00:00Z",
+        plan_name: "plan",
+      }))
+
+      //#when
+      const result = readBoulderState(TEST_DIR)
+
+      //#then
+      expect(result).not.toBeNull()
+      expect(result!.session_ids).toEqual([])
+    })
+
+    test("should default session_ids to [] when not an array", () => {
+      //#given - boulder.json with session_ids as a string
+      const boulderFile = join(SISYPHUS_DIR, "boulder.json")
+      writeFileSync(boulderFile, JSON.stringify({
+        active_plan: "/path/to/plan.md",
+        started_at: "2026-01-01T00:00:00Z",
+        session_ids: "not-an-array",
+        plan_name: "plan",
+      }))
+
+      //#when
+      const result = readBoulderState(TEST_DIR)
+
+      //#then
+      expect(result).not.toBeNull()
+      expect(result!.session_ids).toEqual([])
+    })
+
+    test("should default session_ids to [] for empty object", () => {
+      //#given - boulder.json with empty object
+      const boulderFile = join(SISYPHUS_DIR, "boulder.json")
+      writeFileSync(boulderFile, JSON.stringify({}))
+
+      //#when
+      const result = readBoulderState(TEST_DIR)
+
+      //#then
+      expect(result).not.toBeNull()
+      expect(result!.session_ids).toEqual([])
+    })
+
     test("should read valid boulder state", () => {
       // given - valid boulder.json
       const state: BoulderState = {
@@ -129,6 +201,23 @@ describe("boulder-state", () => {
       // then
       expect(result).toBeNull()
     })
+
+    test("should not crash when boulder.json has no session_ids field", () => {
+      //#given - boulder.json without session_ids
+      const boulderFile = join(SISYPHUS_DIR, "boulder.json")
+      writeFileSync(boulderFile, JSON.stringify({
+        active_plan: "/plan.md",
+        started_at: "2026-01-01T00:00:00Z",
+        plan_name: "plan",
+      }))
+
+      //#when
+      const result = appendSessionId(TEST_DIR, "ses-new")
+
+      //#then - should not crash and should contain the new session
+      expect(result).not.toBeNull()
+      expect(result!.session_ids).toContain("ses-new")
+    })
   })
 
   describe("clearBoulderState", () => {

src/features/boulder-state/storage.ts

@@ -22,7 +22,14 @@ export function readBoulderState(directory: string): BoulderState | null {
 
   try {
     const content = readFileSync(filePath, "utf-8")
-    return JSON.parse(content) as BoulderState
+    const parsed = JSON.parse(content)
+    if (!parsed || typeof parsed !== "object" || Array.isArray(parsed)) {
+      return null
+    }
+    if (!Array.isArray(parsed.session_ids)) {
+      parsed.session_ids = []
+    }
+    return parsed as BoulderState
   } catch {
     return null
   }
@@ -48,7 +55,10 @@ export function appendSessionId(directory: string, sessionId: string): BoulderSt
   const state = readBoulderState(directory)
   if (!state) return null
 
-  if (!state.session_ids.includes(sessionId)) {
+  if (!state.session_ids?.includes(sessionId)) {
+    if (!Array.isArray(state.session_ids)) {
+      state.session_ids = []
+    }
     state.session_ids.push(sessionId)
     if (writeBoulderState(directory, state)) {
       return state

src/hooks/atlas/event-handler.ts

@@ -41,7 +41,7 @@ export function createAtlasEventHandler(input: {
 
       // Read boulder state FIRST to check if this session is part of an active boulder
       const boulderState = readBoulderState(ctx.directory)
-      const isBoulderSession = boulderState?.session_ids.includes(sessionID) ?? false
+      const isBoulderSession = boulderState?.session_ids?.includes(sessionID) ?? false
 
       const isBackgroundTaskSession = subagentSessions.has(sessionID)
 

src/hooks/atlas/tool-execute-after.ts

@@ -65,7 +65,7 @@ export function createToolExecuteAfterHandler(input: {
       if (boulderState) {
         const progress = getPlanProgress(boulderState.active_plan)
 
-        if (toolInput.sessionID && !boulderState.session_ids.includes(toolInput.sessionID)) {
+        if (toolInput.sessionID && !boulderState.session_ids?.includes(toolInput.sessionID)) {
           appendSessionId(ctx.directory, toolInput.sessionID)
           log(`[${HOOK_NAME}] Appended session to boulder`, {
             sessionID: toolInput.sessionID,

src/hooks/comment-checker/hook.output-guard.test.ts

@@ -0,0 +1,24 @@
+import { describe, it, expect, mock } from "bun:test"
+
+mock.module("./cli-runner", () => ({
+  initializeCommentCheckerCli: () => {},
+  getCommentCheckerCliPathPromise: () => Promise.resolve("/tmp/fake-comment-checker"),
+  isCliPathUsable: () => true,
+  processWithCli: async () => {},
+  processApplyPatchEditsWithCli: async () => {},
+}))
+
+const { createCommentCheckerHooks } = await import("./hook")
+
+describe("comment-checker output guard", () => {
+  //#given output.output is undefined
+  //#when tool.execute.after is called
+  //#then should return without throwing
+  it("should not throw when output.output is undefined", async () => {
+    const hooks = createCommentCheckerHooks()
+    const input = { tool: "Write", sessionID: "ses_test", callID: "call_test" }
+    const output = { title: "ok", output: undefined as unknown as string, metadata: {} }
+
+    await expect(hooks["tool.execute.after"](input, output)).resolves.toBeUndefined()
+  })
+})

src/hooks/comment-checker/hook.ts

@@ -89,6 +89,8 @@ export function createCommentCheckerHooks(config?: CommentCheckerConfig) {
     ): Promise<void> => {
       debugLog("tool.execute.after:", { tool: input.tool, callID: input.callID })
 
+      if (!output.output || typeof output.output !== "string") return
+
       const toolLower = input.tool.toLowerCase()
 
       // Only skip if the output indicates a tool execution failure

src/hooks/edit-error-recovery/hook.ts

@@ -43,6 +43,7 @@ export function createEditErrorRecoveryHook(_ctx: PluginInput) {
       output: { title: string; output: string; metadata: unknown }
     ) => {
       if (input.tool.toLowerCase() !== "edit") return
+      if (!output.output || typeof output.output !== "string") return
 
       const outputLower = output.output.toLowerCase()
       const hasEditError = EDIT_ERROR_PATTERNS.some((pattern) =>

src/hooks/edit-error-recovery/index.test.ts

@@ -21,6 +21,17 @@ describe("createEditErrorRecoveryHook", () => {
       metadata: {},
     })
 
+    describe("#given output.output is undefined", () => {
+      //#when tool.execute.after is called
+      //#then should return without throwing
+      it("#then should not throw", async () => {
+        const input = createInput("Edit")
+        const output = { title: "Edit", output: undefined as unknown as string, metadata: {} }
+
+        await expect(hook["tool.execute.after"](input, output)).resolves.toBeUndefined()
+      })
+    })
+
     describe("#given Edit tool with oldString/newString same error", () => {
       describe("#when the error message is detected", () => {
         it("#then should append the recovery reminder", async () => {

src/hooks/prometheus-md-only/agent-resolution.ts

@@ -43,7 +43,7 @@ export function getAgentFromSession(sessionID: string, directory: string): strin
 
   // Check boulder state (persisted across restarts) - fixes #927
   const boulderState = readBoulderState(directory)
-  if (boulderState?.session_ids.includes(sessionID) && boulderState.agent) {
+  if (boulderState?.session_ids?.includes(sessionID) && boulderState.agent) {
     return boulderState.agent
   }
 

src/hooks/task-resume-info/hook.test.ts

@@ -0,0 +1,17 @@
+import { describe, it, expect } from "bun:test"
+import { createTaskResumeInfoHook } from "./hook"
+
+describe("createTaskResumeInfoHook", () => {
+  describe("tool.execute.after", () => {
+    //#given output.output is undefined
+    //#when tool.execute.after is called
+    //#then should return without throwing
+    it("should not throw when output.output is undefined", async () => {
+      const hook = createTaskResumeInfoHook()
+      const input = { tool: "Task", sessionID: "ses_test", callID: "call_test" }
+      const output = { title: "Result", output: undefined as unknown as string, metadata: {} }
+
+      await expect(hook["tool.execute.after"](input, output)).resolves.toBeUndefined()
+    })
+  })
+})

src/hooks/task-resume-info/hook.ts

@@ -21,6 +21,7 @@ export function createTaskResumeInfoHook() {
     output: { title: string; output: string; metadata: unknown }
   ) => {
     if (!TARGET_TOOLS.includes(input.tool)) return
+    if (!output.output || typeof output.output !== "string") return
     if (output.output.startsWith("Error:") || output.output.startsWith("Failed")) return
     if (output.output.includes("\nto continue:")) return
 

src/tools/look-at/tools.test.ts

@@ -1,4 +1,4 @@
-import { describe, expect, test } from "bun:test"
+import { describe, expect, test, mock } from "bun:test"
 import type { ToolContext } from "@opencode-ai/plugin/tool"
 import { normalizeArgs, validateArgs, createLookAt } from "./tools"
 
@@ -111,16 +111,15 @@ describe("look-at tool", () => {
   })
 
   describe("createLookAt error handling", () => {
-    // given promptAsync throws error
+    // given sync prompt throws and no messages available
     // when LookAt tool executed
-    // then returns error string immediately (no message fetch)
-    test("returns error immediately when promptAsync fails", async () => {
+    // then returns no-response error (fetches messages after catching prompt error)
+    test("returns no-response error when prompt fails and no messages exist", async () => {
       const mockClient = {
         session: {
           get: async () => ({ data: { directory: "/project" } }),
           create: async () => ({ data: { id: "ses_test_prompt_fail" } }),
-          promptAsync: async () => { throw new Error("Network connection failed") },
-          status: async () => ({ data: {} }),
+          prompt: async () => { throw new Error("Network connection failed") },
           messages: async () => ({ data: [] }),
         },
       }
@@ -146,51 +145,10 @@ describe("look-at tool", () => {
         toolContext,
       )
       expect(result).toContain("Error")
-      expect(result).toContain("Network connection failed")
+      expect(result).toContain("multimodal-looker")
     })
 
-    // given promptAsync succeeds but status API fails (polling degrades gracefully)
-    // when LookAt tool executed
-    // then still attempts to fetch messages (graceful degradation)
-    test("fetches messages even when status API fails", async () => {
-      const mockClient = {
-        session: {
-          get: async () => ({ data: { directory: "/project" } }),
-          create: async () => ({ data: { id: "ses_test_poll_timeout" } }),
-          promptAsync: async () => ({}),
-          status: async () => ({ error: new Error("status unavailable") }),
-          messages: async () => ({
-            data: [
-              { info: { role: "assistant", time: { created: 1 } }, parts: [{ type: "text", text: "partial result" }] },
-            ],
-          }),
-        },
-      }
-
-      const tool = createLookAt({
-        client: mockClient,
-        directory: "/project",
-      } as any)
-
-      const toolContext: ToolContext = {
-        sessionID: "parent-session",
-        messageID: "parent-message",
-        agent: "sisyphus",
-        directory: "/project",
-        worktree: "/project",
-        abort: new AbortController().signal,
-        metadata: () => {},
-        ask: async () => {},
-      }
-
-      const result = await tool.execute(
-        { file_path: "/test/file.png", goal: "analyze" },
-        toolContext,
-      )
-      expect(result).toBe("partial result")
-    })
-
-    // given promptAsync succeeds and session becomes idle
+    // given sync prompt succeeds
     // when LookAt tool executed and no assistant message found
     // then returns error about no response
     test("returns error when no assistant message after successful prompt", async () => {
@@ -198,8 +156,7 @@ describe("look-at tool", () => {
         session: {
           get: async () => ({ data: { directory: "/project" } }),
           create: async () => ({ data: { id: "ses_test_no_msg" } }),
-          promptAsync: async () => ({}),
-          status: async () => ({ data: {} }),
+          prompt: async () => ({}),
           messages: async () => ({ data: [] }),
         },
       }
@@ -236,8 +193,7 @@ describe("look-at tool", () => {
         session: {
           get: async () => ({ data: { directory: "/project" } }),
           create: async () => ({ error: "Internal server error" }),
-          promptAsync: async () => ({}),
-          status: async () => ({ data: {} }),
+          prompt: async () => ({}),
           messages: async () => ({ data: [] }),
         },
       }
@@ -270,8 +226,8 @@ describe("look-at tool", () => {
   describe("createLookAt model passthrough", () => {
     // given multimodal-looker agent has resolved model info
     // when LookAt tool executed
-    // then model info should be passed to promptAsync
-    test("passes multimodal-looker model to promptAsync when available", async () => {
+    // then model info should be passed to sync prompt
+    test("passes multimodal-looker model to sync prompt when available", async () => {
       let promptBody: any
 
       const mockClient = {
@@ -289,11 +245,10 @@ describe("look-at tool", () => {
         session: {
           get: async () => ({ data: { directory: "/project" } }),
           create: async () => ({ data: { id: "ses_model_passthrough" } }),
-          promptAsync: async (input: any) => {
+          prompt: async (input: any) => {
             promptBody = input.body
             return { data: {} }
           },
-          status: async () => ({ data: {} }),
           messages: async () => ({
             data: [
               { info: { role: "assistant", time: { created: 1 } }, parts: [{ type: "text", text: "done" }] },
@@ -330,10 +285,154 @@ describe("look-at tool", () => {
     })
   })
 
+  describe("createLookAt sync prompt (race condition fix)", () => {
+    // given look_at needs response immediately after prompt returns
+    // when tool is executed
+    // then must use synchronous prompt (session.prompt), NOT async (session.promptAsync)
+    test("uses synchronous prompt to avoid race condition with polling", async () => {
+      const syncPrompt = mock(async () => ({}))
+      const asyncPrompt = mock(async () => ({}))
+      const statusFn = mock(async () => ({ data: {} }))
+
+      const mockClient = {
+        app: {
+          agents: async () => ({ data: [] }),
+        },
+        session: {
+          get: async () => ({ data: { directory: "/project" } }),
+          create: async () => ({ data: { id: "ses_sync_test" } }),
+          prompt: syncPrompt,
+          promptAsync: asyncPrompt,
+          status: statusFn,
+          messages: async () => ({
+            data: [
+              { info: { role: "assistant", time: { created: 1 } }, parts: [{ type: "text", text: "result" }] },
+            ],
+          }),
+        },
+      }
+
+      const tool = createLookAt({
+        client: mockClient,
+        directory: "/project",
+      } as any)
+
+      const toolContext: ToolContext = {
+        sessionID: "parent-session",
+        messageID: "parent-message",
+        agent: "sisyphus",
+        directory: "/project",
+        worktree: "/project",
+        abort: new AbortController().signal,
+        metadata: () => {},
+        ask: async () => {},
+      }
+
+      const result = await tool.execute(
+        { file_path: "/test/file.png", goal: "analyze" },
+        toolContext,
+      )
+
+      expect(result).toBe("result")
+      expect(syncPrompt).toHaveBeenCalledTimes(1)
+      expect(asyncPrompt).not.toHaveBeenCalled()
+      expect(statusFn).not.toHaveBeenCalled()
+    })
+
+    // given sync prompt throws (JSON parse error even on success)
+    // when tool is executed
+    // then catches error gracefully and still fetches messages
+    test("catches sync prompt errors and still fetches messages", async () => {
+      const mockClient = {
+        app: {
+          agents: async () => ({ data: [] }),
+        },
+        session: {
+          get: async () => ({ data: { directory: "/project" } }),
+          create: async () => ({ data: { id: "ses_sync_error" } }),
+          prompt: async () => { throw new Error("JSON parse error") },
+          promptAsync: async () => ({}),
+          status: async () => ({ data: {} }),
+          messages: async () => ({
+            data: [
+              { info: { role: "assistant", time: { created: 1 } }, parts: [{ type: "text", text: "result despite error" }] },
+            ],
+          }),
+        },
+      }
+
+      const tool = createLookAt({
+        client: mockClient,
+        directory: "/project",
+      } as any)
+
+      const toolContext: ToolContext = {
+        sessionID: "parent-session",
+        messageID: "parent-message",
+        agent: "sisyphus",
+        directory: "/project",
+        worktree: "/project",
+        abort: new AbortController().signal,
+        metadata: () => {},
+        ask: async () => {},
+      }
+
+      const result = await tool.execute(
+        { file_path: "/test/file.png", goal: "analyze" },
+        toolContext,
+      )
+
+      expect(result).toBe("result despite error")
+    })
+
+    // given sync prompt throws and no messages available
+    // when tool is executed
+    // then returns error about no response
+    test("returns no-response error when sync prompt fails and no messages", async () => {
+      const mockClient = {
+        app: {
+          agents: async () => ({ data: [] }),
+        },
+        session: {
+          get: async () => ({ data: { directory: "/project" } }),
+          create: async () => ({ data: { id: "ses_sync_no_msg" } }),
+          prompt: async () => { throw new Error("Connection refused") },
+          promptAsync: async () => ({}),
+          status: async () => ({ data: {} }),
+          messages: async () => ({ data: [] }),
+        },
+      }
+
+      const tool = createLookAt({
+        client: mockClient,
+        directory: "/project",
+      } as any)
+
+      const toolContext: ToolContext = {
+        sessionID: "parent-session",
+        messageID: "parent-message",
+        agent: "sisyphus",
+        directory: "/project",
+        worktree: "/project",
+        abort: new AbortController().signal,
+        metadata: () => {},
+        ask: async () => {},
+      }
+
+      const result = await tool.execute(
+        { file_path: "/test/file.png", goal: "analyze" },
+        toolContext,
+      )
+
+      expect(result).toContain("Error")
+      expect(result).toContain("multimodal-looker")
+    })
+  })
+
   describe("createLookAt with image_data", () => {
     // given base64 image data is provided
     // when LookAt tool executed
-    // then should send data URL to promptAsync
+    // then should send data URL to sync prompt
     test("sends data URL when image_data provided", async () => {
       let promptBody: any
 
@@ -344,11 +443,10 @@ describe("look-at tool", () => {
         session: {
           get: async () => ({ data: { directory: "/project" } }),
           create: async () => ({ data: { id: "ses_image_data_test" } }),
-          promptAsync: async (input: any) => {
+          prompt: async (input: any) => {
             promptBody = input.body
             return { data: {} }
           },
-          status: async () => ({ data: {} }),
           messages: async () => ({
             data: [
               { info: { role: "assistant", time: { created: 1 } }, parts: [{ type: "text", text: "analyzed" }] },
@@ -398,11 +496,10 @@ describe("look-at tool", () => {
         session: {
           get: async () => ({ data: { directory: "/project" } }),
           create: async () => ({ data: { id: "ses_raw_base64_test" } }),
-          promptAsync: async (input: any) => {
+          prompt: async (input: any) => {
             promptBody = input.body
             return { data: {} }
           },
-          status: async () => ({ data: {} }),
           messages: async () => ({
             data: [
               { info: { role: "assistant", time: { created: 1 } }, parts: [{ type: "text", text: "analyzed" }] },

src/tools/look-at/tools.ts

@@ -3,8 +3,7 @@ import { pathToFileURL } from "node:url"
 import { tool, type PluginInput, type ToolDefinition } from "@opencode-ai/plugin"
 import { LOOK_AT_DESCRIPTION, MULTIMODAL_LOOKER_AGENT } from "./constants"
 import type { LookAtArgs } from "./types"
-import { log, promptWithModelSuggestionRetry } from "../../shared"
-import { pollSessionUntilIdle } from "./session-poller"
+import { log, promptSyncWithModelSuggestionRetry } from "../../shared"
 import { extractLatestAssistantText } from "./assistant-message-extractor"
 import type { LookAtArgsWithAlias } from "./look-at-arguments"
 import { normalizeArgs, validateArgs } from "./look-at-arguments"
@@ -106,9 +105,9 @@ Original error: ${createResult.error}`
 
       const { agentModel, agentVariant } = await resolveMultimodalLookerAgentMetadata(ctx)
 
-      log(`[look_at] Sending async prompt with ${isBase64Input ? "base64 image" : "file"} to session ${sessionID}`)
+      log(`[look_at] Sending prompt with ${isBase64Input ? "base64 image" : "file"} to session ${sessionID}`)
       try {
-        await promptWithModelSuggestionRetry(ctx.client, {
+        await promptSyncWithModelSuggestionRetry(ctx.client, {
           path: { id: sessionID },
           body: {
             agent: MULTIMODAL_LOOKER_AGENT,
@@ -127,15 +126,7 @@ Original error: ${createResult.error}`
           },
         })
       } catch (promptError) {
-        log(`[look_at] promptAsync error:`, promptError)
-        return `Error: Failed to send prompt to multimodal-looker agent: ${promptError instanceof Error ? promptError.message : String(promptError)}`
-      }
-
-      log(`[look_at] Polling session ${sessionID} until idle...`)
-      try {
-        await pollSessionUntilIdle(ctx.client, sessionID, { pollIntervalMs: 500, timeoutMs: 120_000 })
-      } catch (pollError) {
-        log(`[look_at] Polling error (will still try to fetch messages):`, pollError)
+        log(`[look_at] Prompt error (ignored, will still fetch messages):`, promptError)
       }
 
       log(`[look_at] Fetching messages from session ${sessionID}...`)