@elvatis_com/openclaw-cli-bridge-elvatis 3.4.1 → 3.5.0

package/README.md

@@ -2,7 +2,7 @@
 
 > OpenClaw plugin that bridges locally installed AI CLIs (Codex, Gemini, Claude Code, OpenCode, Pi) as model providers — with slash commands for instant model switching, restore, health testing, and model listing.
 
-**Current version:** `3.4.1`
+**Current version:** `3.5.0`
 
 ---
 

package/SKILL.md

@@ -68,4 +68,4 @@ On gateway restart, if any session has expired, a **WhatsApp alert** is sent aut
 
 See `README.md` for full configuration reference and architecture diagram.
 
-**Version:** 3.4.1
+**Version:** 3.5.0

package/openclaw.plugin.json

@@ -2,7 +2,7 @@
   "id": "openclaw-cli-bridge-elvatis",
   "slug": "openclaw-cli-bridge-elvatis",
   "name": "OpenClaw CLI Bridge",
-  "version": "3.4.1",
+  "version": "3.5.0",
   "license": "MIT",
   "description": "Phase 1: openai-codex auth bridge. Phase 2: local HTTP proxy routing model calls through gemini/claude CLIs (vllm provider).",
   "providers": [

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@elvatis_com/openclaw-cli-bridge-elvatis",
-  "version": "3.4.1",
+  "version": "3.5.0",
   "description": "Bridges gemini, claude, and codex CLI tools as OpenClaw model providers. Reads existing CLI auth without re-login.",
   "type": "module",
   "openclaw": {

package/src/cli-runner.ts

@@ -80,11 +80,16 @@ export function formatPrompt(messages: ChatMessage[], toolCount = 0): string {
   // Reduce history when tool schemas dominate the prompt
   const maxMsgs = toolCount > TOOL_HEAVY_THRESHOLD ? MAX_MESSAGES_HEAVY_TOOLS : MAX_MESSAGES;
 
-  // Keep system message (if any) + last N non-system messages
+  // Keep system message (if any) + first user message (original request) + last N non-system messages
   const system = messages.find((m) => m.role === "system");
   const nonSystem = messages.filter((m) => m.role !== "system");
+  const firstUser = nonSystem.find((m) => m.role === "user");
   const recent = nonSystem.slice(-maxMsgs);
-  const truncated = system ? [system, ...recent] : recent;
+  // Pin the first user message so the model never loses the original request
+  const pinned = firstUser && !recent.includes(firstUser)
+    ? [firstUser, ...recent]
+    : recent;
+  const truncated = system ? [system, ...pinned] : pinned;
 
   // Single short user message — send bare (no wrapping needed)
   if (truncated.length === 1 && truncated[0].role === "user") {

package/test/cli-runner.test.ts

@@ -19,18 +19,19 @@ describe("formatPrompt", () => {
     expect(result).toBe("hello");
   });
 
-  it("truncates to MAX_MESSAGES (20) non-system messages", () => {
+  it("truncates to MAX_MESSAGES (20) non-system messages but pins first user message", () => {
     const messages = Array.from({ length: 30 }, (_, i) => ({
       role: "user" as const,
       content: `msg ${i}`,
     }));
     const result = formatPrompt(messages);
     expect(result).toContain("msg 29");
-    expect(result).not.toContain("msg 0\n");
+    expect(result).toContain("msg 0"); // first user message is always pinned
+    expect(result).not.toContain("msg 1\n"); // but intermediate messages are truncated
     expect(result).toContain("[User]");
   });
 
-  it("keeps system message + last 20 non-system messages", () => {
+  it("keeps system message + first user message + last 20 non-system messages", () => {
     const sys = { role: "system" as const, content: "You are helpful" };
     const msgs = Array.from({ length: 25 }, (_, i) => ({
       role: "user" as const,
@@ -40,7 +41,8 @@ describe("formatPrompt", () => {
     expect(result).toContain("[System]");
     expect(result).toContain("You are helpful");
     expect(result).toContain("msg 24");
-    expect(result).not.toContain("msg 0\n");
+    expect(result).toContain("msg 0"); // first user message is always pinned
+    expect(result).not.toContain("msg 1\n"); // but intermediate messages are truncated
   });
 
   it("truncates individual message content at MAX_MSG_CHARS (4000)", () => {