Merge pull request #4171 from MoerAI/fix/multimodal-looker-tool-usage-guidance

fix(agents): declare multimodal-looker tool allowlist in prompt to prevent death loop on small VL models (fixes #4116)

src/agents/multimodal-looker.test.ts

@@ -0,0 +1,27 @@
+import { describe, test, expect } from "bun:test"
+import { createMultimodalLookerAgent } from "./multimodal-looker"
+
+describe("createMultimodalLookerAgent", () => {
+  test("prompt explicitly enumerates the agent's available tools to prevent death loop on small VL models", () => {
+    // given
+    const agent = createMultimodalLookerAgent("openai/gpt-5-nano")
+
+    // when
+    const prompt = typeof agent.prompt === "string" ? agent.prompt : ""
+
+    // then
+    expect(prompt).toMatch(/available tools/i)
+    expect(prompt).toContain("read")
+  })
+
+  test("prompt instructs the agent never to call other tools", () => {
+    // given
+    const agent = createMultimodalLookerAgent("openai/gpt-5-nano")
+
+    // when
+    const prompt = typeof agent.prompt === "string" ? agent.prompt : ""
+
+    // then
+    expect(prompt.toLowerCase()).toContain("never")
+  })
+})

src/agents/multimodal-looker.ts

@@ -23,6 +23,8 @@ export function createMultimodalLookerAgent(model: string): AgentConfig {
     ...restrictions,
     prompt: `You interpret media files that cannot be read as plain text.
 
+Your only available tools are 'read' and 'call_omo_agent'. Always use 'read' to load the file first, then analyze the returned content. Never attempt to call any other tool.
+
 Your job: examine the attached file and extract ONLY what was requested.
 
 When to use you: