@elvatis_com/openclaw-cli-bridge-elvatis 1.6.2 → 1.6.4

package/.ai/handoff/STATUS.md

@@ -1,46 +1,45 @@
 # STATUS — openclaw-cli-bridge-elvatis
 
-## Current Version: 1.4.0
+## Current Version: 1.6.3
 
-## All 4 Providers Available — on-demand via /xxx-login (no CDP required)
-| Provider | Status | Models | Command |
-|---|---|---|---|
-| Grok | ✅ | web-grok/grok-3, grok-3-fast, grok-3-mini, grok-3-mini-fast | /grok-login |
-| Claude | ✅ | web-claude/claude-sonnet, claude-opus, claude-haiku | /claude-login |
-| Gemini | ✅ | web-gemini/gemini-2-5-pro, gemini-2-5-flash, gemini-3-pro, gemini-3-flash | /gemini-login |
-| ChatGPT | ✅ | web-chatgpt/gpt-4o, gpt-4o-mini, gpt-o3, gpt-o4-mini, gpt-5 | /chatgpt-login |
+## All 4 Providers Available — persistent Chromium profiles
+| Provider | Status | Models | Login Cmd | Cookie Expiry |
+|---|---|---|---|---|
+| Grok | ✅ | web-grok/grok-3, grok-3-fast, grok-3-mini, grok-3-mini-fast | /grok-login | ~178d |
+| Gemini | ✅ | web-gemini/gemini-2-5-pro, gemini-2-5-flash, gemini-3-pro, gemini-3-flash | /gemini-login | ~398d |
+| Claude.ai | ⚠️ expired | web-claude/claude-sonnet, claude-opus, claude-haiku | /claude-login | EXPIRED |
+| ChatGPT | ⚠️ expiring | web-chatgpt/gpt-4o, gpt-4o-mini, gpt-4.1, o3, o4-mini, gpt-5, gpt-5-mini | /chatgpt-login | ~6d |
 
 ## Stats
-- 22 total models, 16 web-session models
+- 22 total models (6 CLI + 16 web-session)
 - 96/96 tests green (8 test files)
-- 0 zombie Chromium processes at startup (browsers on-demand only)
-- Cookie expiry tracking for all 4 providers
-- Singleton guard on ensureAllProviderContexts (no concurrent spawns)
-- Persistent Chromium fallback for all 4 providers (no CDP dependency)
+- All 4 providers use launchPersistentContext — sessions survive gateway restarts
+- /bridge-status shows cookie-based status (independent of in-memory context)
 
 ## Architecture: Browser Lifecycle
-- **On plugin start:** NO browser launched automatically
-- **On /xxx-login:** tries CDP import (optional) then launches persistent Chromium for that provider
-- **CDP unavailable?** No problem - persistent Chromium from saved profile is used automatically
-- **On request (no context):** proxy callbacks also use persistent fallback before returning null
-- **On /xxx-logout:** closes context + deletes session file
-
-## Known Issues
-- Cloudflare may block headless Chromium for some providers without cookies
-  - Workaround: first login via OpenClaw browser (CDP), subsequent restarts use saved profile
+- **Profile dirs:** `~/.openclaw/{grok,gemini,claude,chatgpt}-profile/`
+- **On plugin start:** startup restore attempts headless reconnect from saved profiles (5s delay)
+- **On /xxx-login:** headed browser, user logs in, cookies baked to profile automatically
+- **On request (no in-memory ctx):** proxy lazy-launches persistent context on first request
+- **On /xxx-logout:** closes context + deletes profile + clears expiry file
+- **bridge-status:** uses cookie expiry files as source of truth (not in-memory state)
+  - ✅ active — browser connected and verified
+  - 🟡 logged in, browser not loaded — cookies valid, lazy-loads on first request
+  - 🔴 session expired — needs /xxx-login
+  - ⚪ never logged in
 
 ## Release History
+- v1.6.1 (2026-03-13): Fix /bridge-status — use cookie expiry as source of truth, not in-memory context
+- v1.6.0 (2026-03-13): Persistent Chromium profiles for all 4 providers (Claude web + ChatGPT)
+- v1.5.1 (2026-03-12): Fix hardcoded plugin version
+- v1.5.0 (2026-03-12): Remove /claude-login and /chatgpt-login (pre-v1.6.0 interim)
 - v1.4.0 (2026-03-12): Persistent browser fallback for Claude/Gemini/ChatGPT (no CDP required)
-- v1.3.5 (2026-03-12): Startup restore guard - runs only once per process (SIGUSR1 fix)
-- v1.3.4 (2026-03-12): Safe sequential session restore from saved profiles
-- v1.3.3 (2026-03-12): Remove startup auto-connect - browsers on-demand only (OOM fix)
-- v1.3.2 (2026-03-12): Singleton guard on ensureAllProviderContexts (resource leak fix)
-- v1.3.1 (2026-03-11): Cookie baking into persistent profiles on login
+- v1.3.5 (2026-03-12): Startup restore guard (SIGUSR1 OOM fix)
 - v1.3.0 (2026-03-11): Browser auto-reconnect after gateway restart
-- v1.2.0 (2026-03-11): Fresh page per request + ChatGPT model switching
-- v1.1.0 (2026-03-11): Auto-connect on startup + /bridge-status
-- v1.0.0 (2026-03-11): All 4 providers headless (Grok/Claude/Gemini/ChatGPT) - 96/96 tests
+- v1.0.0 (2026-03-11): All 4 providers headless — 96/96 tests
 
 ## Next Steps
+- /claude-login needs to be run (session expired)
+- /chatgpt-login needs to be run in ~6 days
+- Gemini model switching via UI (2.5 Pro vs Flash vs 3)
 - Context-window management for long conversations
-- Gemini model switching (2.5 Pro vs Flash vs 3) via UI

package/README.md

@@ -2,7 +2,7 @@
 
 > OpenClaw plugin that bridges locally installed AI CLIs (Codex, Gemini, Claude Code) as model providers — with slash commands for instant model switching, restore, health testing, and model listing.
 
-**Current version:** `1.6.2`
+**Current version:** `1.6.4`
 
 ---
 
@@ -135,9 +135,11 @@ Routes requests through real browser sessions on the provider's web UI. Requires
 |---|---|
 | `web-chatgpt/gpt-4o` | GPT-4o |
 | `web-chatgpt/gpt-4o-mini` | GPT-4o Mini |
-| `web-chatgpt/gpt-o3` | GPT o3 |
-| `web-chatgpt/gpt-o4-mini` | GPT o4-mini |
+| `web-chatgpt/gpt-4.1` | GPT-4.1 |
+| `web-chatgpt/o3` | o3 |
+| `web-chatgpt/o4-mini` | o4-mini |
 | `web-chatgpt/gpt-5` | GPT-5 |
+| `web-chatgpt/gpt-5-mini` | GPT-5 Mini |
 
 | Command | What it does |
 |---|---|
@@ -360,6 +362,14 @@ npm test            # vitest run (83 tests)
 
 ## Changelog
 
+### v1.6.4
+- **chore:** version bump (1.6.3 was already published on npm with partial changes)
+
+### v1.6.3
+- **fix:** CLI-proxy models (`cli-gemini/*`, `cli-claude/*`) now return HTTP 400 with `tools_not_supported` when a request includes tool/function call schemas — prevents agents from silently failing or hallucinating when assigned a CLI-proxy model
+- **feat:** `/v1/models` response includes `capabilities.tools: false` for CLI-proxy models so OpenClaw can detect tool support upfront
+- **fix:** EADDRINUSE on hot-reload: re-probe after 1.5s wait before retrying bind; probe timeout 800ms → 2000ms
+
 ### v1.6.2
 - **docs:** Add missing changelog entries (v1.5.1, v1.6.0, v1.6.1), fix /cli-codex54 command name in SKILL.md, add startup re-login alert description to SKILL.md.
 
@@ -458,7 +468,7 @@ No CLI binaries required — just authenticated browser sessions.
 
 ## v1.0.0
 - **feat:** `chatgpt-browser.ts` — chatgpt.com DOM-automation (`#prompt-textarea` + `[data-message-author-role]`)
-- **feat:** `web-chatgpt/*` models: gpt-4o, gpt-4o-mini, gpt-o3, gpt-o4-mini, gpt-5
+- **feat:** `web-chatgpt/*` models: gpt-4o, gpt-4o-mini, gpt-4.1, o3, o4-mini, gpt-5, gpt-5-mini (updated in v1.6.3)
 - **feat:** `/chatgpt-login`, `/chatgpt-status`, `/chatgpt-logout` + cookie-expiry tracking
 - **feat:** All 4 providers headless: Grok ✅ Claude ✅ Gemini ✅ ChatGPT ✅
 - **test:** 96/83 tests green (8 test files)

package/SKILL.md

@@ -64,4 +64,4 @@ 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:** 1.6.2
+**Version:** 1.6.4

package/index.ts

@@ -854,7 +854,7 @@ function proxyTestRequest(
 const plugin = {
   id: "openclaw-cli-bridge-elvatis",
   name: "OpenClaw CLI Bridge",
-  version: "1.6.2",
+  version: "1.6.4",
   description:
     "Phase 1: openai-codex auth bridge. " +
     "Phase 2: HTTP proxy for gemini/claude CLIs. " +
@@ -1067,7 +1067,7 @@ const plugin = {
               headers: { Authorization: `Bearer ${apiKey}` } },
             (res) => { res.resume(); resolve(res.statusCode === 200); }
           );
-          req.setTimeout(800, () => { req.destroy(); resolve(false); });
+          req.setTimeout(2000, () => { req.destroy(); resolve(false); });
           req.on("error", () => resolve(false));
           req.end();
         });
@@ -1144,8 +1144,17 @@ const plugin = {
         } catch (err: unknown) {
           const msg = (err as Error).message ?? String(err);
           if (msg.includes("EADDRINUSE")) {
-            // Port is busy but probe didn't respond — wait for the OS to release it
-            api.logger.warn(`[cli-bridge] port ${port} busy, waiting 1s for OS release…`);
+            // Port is busy but probe didn't respond — maybe the old server is still shutting down.
+            // Re-probe first: if it now responds, reuse it without rebinding.
+            api.logger.warn(`[cli-bridge] port ${port} busy — re-probing before retry…`);
+            await new Promise(r => setTimeout(r, 1500));
+            const aliveNow = await probeExisting();
+            if (aliveNow) {
+              api.logger.info(`[cli-bridge] proxy now responding on :${port} — reusing`);
+              return;
+            }
+            // Still not responding — wait for OS to release the port, then rebind
+            api.logger.warn(`[cli-bridge] port ${port} still busy, waiting 1s for OS release…`);
             await new Promise((r) => setTimeout(r, 1000));
             // One final attempt
             try {
@@ -1975,22 +1984,41 @@ const plugin = {
                 return page.locator("#prompt-textarea").isVisible().catch(() => false);
               } catch { chatgptContext = null; return false; }
             },
-            models: "web-chatgpt/gpt-4o, gpt-4o-mini, gpt-o3, gpt-o4-mini, gpt-5",
+            models: "web-chatgpt/gpt-4o, gpt-4o-mini, gpt-4.1, o3, o4-mini, gpt-5, gpt-5-mini",
             loginCmd: "/chatgpt-login",
             expiry: () => { const e = loadChatGPTExpiry(); return e ? formatChatGPTExpiry(e) : null; },
           },
         ];
 
         for (const c of checks) {
-          const active = c.ctx !== null;
-          const ok = active ? await c.check() : false;
           const expiry = c.expiry();
-          if (ok) {
-            lines.push(`✅ *${c.name}* — active`);
+          const inMemory = c.ctx !== null;
+          const liveOk = inMemory ? await c.check() : false;
+
+          // Cookie-based status: treat as "logged in" if expiry file exists and not expired
+          // ⚠️ EXPIRED = truly expired, 🚨 = expiring soon (still valid), ✅ = fine
+          // This reflects actual login state independent of in-memory context
+          const cookieExpired = expiry !== null && expiry.startsWith("⚠️ EXPIRED");
+          const cookieValid = expiry !== null && !cookieExpired;
+
+          if (liveOk) {
+            // In-memory context active and verified
+            lines.push(`✅ *${c.name}* — active (browser connected)`);
             if (expiry) lines.push(`   🕐 ${expiry}`);
             lines.push(`   Models: ${c.models}`);
+          } else if (cookieValid) {
+            // Not in memory yet, but cookies are valid — will auto-connect on next request
+            lines.push(`🟡 *${c.name}* — logged in, browser not loaded`);
+            lines.push(`   🕐 ${expiry}`);
+            lines.push(`   Models: ${c.models}`);
+            lines.push(`   ℹ️ Browser launches on first request`);
+          } else if (cookieExpired) {
+            // Cookies expired — needs re-login
+            lines.push(`🔴 *${c.name}* — session expired (run \`${c.loginCmd}\`)`);
+            if (expiry) lines.push(`   🕐 ${expiry}`);
           } else {
-            lines.push(`❌ *${c.name}* — not connected (run \`${c.loginCmd}\`)`);
+            // No cookie file at all — never logged in
+            lines.push(`⚪ *${c.name}* — never logged in (run \`${c.loginCmd}\`)`);
           }
           lines.push("");
         }

package/openclaw.plugin.json

@@ -1,7 +1,7 @@
 {
   "id": "openclaw-cli-bridge-elvatis",
   "name": "OpenClaw CLI Bridge",
-  "version": "1.6.2",
+  "version": "1.6.4",
   "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": "1.6.2",
+  "version": "1.6.4",
   "description": "Bridges gemini, claude, and codex CLI tools as OpenClaw model providers. Reads existing CLI auth without re-login.",
   "type": "module",
   "openclaw": {

package/src/chatgpt-browser.ts

@@ -38,18 +38,26 @@ const CHATGPT_HOME = "https://chatgpt.com";
 const MODEL_URLS: Record<string, string> = {
   "gpt-4o":       "https://chatgpt.com/?model=gpt-4o",
   "gpt-4o-mini":  "https://chatgpt.com/?model=gpt-4o-mini",
+  "gpt-4.1":      "https://chatgpt.com/?model=gpt-4.1",
   "o3":           "https://chatgpt.com/?model=o3",
   "o4-mini":      "https://chatgpt.com/?model=o4-mini",
   "gpt-5":        "https://chatgpt.com/?model=gpt-5",
+  "gpt-5-mini":   "https://chatgpt.com/?model=gpt-5-mini",
 };
 
 const MODEL_MAP: Record<string, string> = {
+  // canonical names
   "gpt-4o":        "gpt-4o",
   "gpt-4o-mini":   "gpt-4o-mini",
+  "gpt-4.1":       "gpt-4.1",
+  "o3":            "o3",
+  "o4-mini":       "o4-mini",
+  "gpt-5":         "gpt-5",
+  "gpt-5-mini":    "gpt-5-mini",
+  // aliases (web-chatgpt/* prefix stripped by proxy)
   "gpt-o3":        "o3",
   "gpt-o4-mini":   "o4-mini",
   "gpt-4-1":       "gpt-4.1",
-  "gpt-5":         "gpt-5",
 };
 
 function resolveModel(m?: string): string {

package/src/cli-runner.ts

@@ -316,8 +316,20 @@ export const DEFAULT_ALLOWED_CLI_MODELS: ReadonlySet<string> = new Set([
   "cli-gemini/gemini-2.5-flash",
   "cli-gemini/gemini-3-pro-preview",
   "cli-gemini/gemini-3-flash-preview",
+  // Aliases (map to preview variants internally)
+  "cli-gemini/gemini-3-pro",   // alias → gemini-3-pro-preview
+  "cli-gemini/gemini-3-flash", // alias → gemini-3-flash-preview
 ]);
 
+/** Normalize model aliases to their canonical CLI model names. */
+function normalizeModelAlias(normalized: string): string {
+  const ALIASES: Record<string, string> = {
+    "cli-gemini/gemini-3-pro":   "cli-gemini/gemini-3-pro-preview",
+    "cli-gemini/gemini-3-flash": "cli-gemini/gemini-3-flash-preview",
+  };
+  return ALIASES[normalized] ?? normalized;
+}
+
 // ──────────────────────────────────────────────────────────────────────────────
 // Router
 // ──────────────────────────────────────────────────────────────────────────────
@@ -364,8 +376,11 @@ export async function routeToCliRunner(
     );
   }
 
-  if (normalized.startsWith("cli-gemini/")) return runGemini(prompt, normalized, timeoutMs);
-  if (normalized.startsWith("cli-claude/")) return runClaude(prompt, normalized, timeoutMs);
+  // Resolve aliases (e.g. gemini-3-pro → gemini-3-pro-preview) after allowlist check
+  const resolved = normalizeModelAlias(normalized);
+
+  if (resolved.startsWith("cli-gemini/")) return runGemini(prompt, resolved, timeoutMs);
+  if (resolved.startsWith("cli-claude/")) return runClaude(prompt, resolved, timeoutMs);
 
   throw new Error(
     `Unknown CLI bridge model: "${model}". Use "vllm/cli-gemini/<model>" or "vllm/cli-claude/<model>".`

package/src/proxy-server.ts

@@ -72,6 +72,7 @@ export const CLI_MODELS = [
   { id: "cli-gemini/gemini-2.5-pro",      name: "Gemini 2.5 Pro (CLI)",   contextWindow: 1_000_000, maxTokens: 8_192 },
   { id: "cli-gemini/gemini-2.5-flash",    name: "Gemini 2.5 Flash (CLI)", contextWindow: 1_000_000, maxTokens: 8_192 },
   { id: "cli-gemini/gemini-3-pro-preview",name: "Gemini 3 Pro (CLI)",     contextWindow: 1_000_000, maxTokens: 8_192 },
+  { id: "cli-gemini/gemini-3-pro",        name: "Gemini 3 Pro (CLI, alias)", contextWindow: 1_000_000, maxTokens: 8_192 },
   // Grok web-session models (requires /grok-login)
   { id: "web-grok/grok-3",           name: "Grok 3 (web session)",           contextWindow: 131_072, maxTokens: 131_072 },
   { id: "web-grok/grok-3-fast",      name: "Grok 3 Fast (web session)",      contextWindow: 131_072, maxTokens: 131_072 },
@@ -87,11 +88,13 @@ export const CLI_MODELS = [
   { id: "web-claude/claude-opus",       name: "Claude Opus (web session)",       contextWindow: 200_000, maxTokens: 8192 },
   { id: "web-claude/claude-haiku",      name: "Claude Haiku (web session)",      contextWindow: 200_000, maxTokens: 8192 },
   // ChatGPT web-session models (requires /chatgpt-login)
-  { id: "web-chatgpt/gpt-4o",           name: "GPT-4o (web session)",            contextWindow: 128_000, maxTokens: 8192 },
-  { id: "web-chatgpt/gpt-4o-mini",      name: "GPT-4o Mini (web session)",       contextWindow: 128_000, maxTokens: 8192 },
-  { id: "web-chatgpt/gpt-o3",           name: "GPT o3 (web session)",            contextWindow: 200_000, maxTokens: 8192 },
-  { id: "web-chatgpt/gpt-o4-mini",      name: "GPT o4-mini (web session)",       contextWindow: 200_000, maxTokens: 8192 },
-  { id: "web-chatgpt/gpt-5",            name: "GPT-5 (web session)",             contextWindow: 200_000, maxTokens: 8192 },
+  { id: "web-chatgpt/gpt-4o",           name: "GPT-4o (web session)",            contextWindow: 128_000, maxTokens: 16_384 },
+  { id: "web-chatgpt/gpt-4o-mini",      name: "GPT-4o Mini (web session)",       contextWindow: 128_000, maxTokens: 16_384 },
+  { id: "web-chatgpt/gpt-4.1",          name: "GPT-4.1 (web session)",           contextWindow: 1_047_576, maxTokens: 32_768 },
+  { id: "web-chatgpt/o3",               name: "o3 (web session)",                contextWindow: 200_000, maxTokens: 100_000 },
+  { id: "web-chatgpt/o4-mini",          name: "o4-mini (web session)",           contextWindow: 200_000, maxTokens: 100_000 },
+  { id: "web-chatgpt/gpt-5",            name: "GPT-5 (web session)",             contextWindow: 1_047_576, maxTokens: 32_768 },
+  { id: "web-chatgpt/gpt-5-mini",       name: "GPT-5 Mini (web session)",        contextWindow: 1_047_576, maxTokens: 32_768 },
 ];
 
 // ──────────────────────────────────────────────────────────────────────────────
@@ -165,6 +168,10 @@ async function handleRequest(
           object: "model",
           created: now,
           owned_by: "openclaw-cli-bridge",
+          // CLI-proxy models stream plain text — no tool/function call support
+          capabilities: {
+            tools: !(m.id.startsWith("cli-gemini/") || m.id.startsWith("cli-claude/")),
+          },
         })),
       })
     );
@@ -199,7 +206,8 @@ async function handleRequest(
       return;
     }
 
-    const { model, messages, stream = false } = parsed;
+    const { model, messages, stream = false } = parsed as { model: string; messages: ChatMessage[]; stream?: boolean; tools?: unknown };
+    const hasTools = Array.isArray((parsed as { tools?: unknown }).tools) && (parsed as { tools?: unknown[] }).tools!.length > 0;
 
     if (!model || !messages?.length) {
       res.writeHead(400, { "Content-Type": "application/json" });
@@ -207,7 +215,23 @@ async function handleRequest(
       return;
     }
 
-    opts.log(`[cli-bridge] ${model} · ${messages.length} msg(s) · stream=${stream}`);
+    // CLI-proxy models (cli-gemini/*, cli-claude/*) are plain text completions —
+    // they cannot process tool/function call schemas. Return a clear 400 so
+    // OpenClaw can surface a meaningful error instead of getting a garbled response.
+    const isCliModel = model.startsWith("cli-gemini/") || model.startsWith("cli-claude/");
+    if (hasTools && isCliModel) {
+      res.writeHead(400, { "Content-Type": "application/json" });
+      res.end(JSON.stringify({
+        error: {
+          message: `Model ${model} does not support tool/function calls. Use a native API model (e.g. github-copilot/gpt-5-mini) for agents that need tools.`,
+          type: "invalid_request_error",
+          code: "tools_not_supported",
+        }
+      }));
+      return;
+    }
+
+    opts.log(`[cli-bridge] ${model} · ${messages.length} msg(s) · stream=${stream}${hasTools ? " · tools=unsupported→rejected" : ""}`);
 
     const id = `chatcmpl-cli-${randomBytes(6).toString("hex")}`;
     const created = Math.floor(Date.now() / 1000);

package/test/chatgpt-proxy.test.ts

@@ -81,14 +81,16 @@ describe("ChatGPT web-session routing — model list", () => {
     const ids = (res.body as { data: { id: string }[] }).data.map(m => m.id);
     expect(ids).toContain("web-chatgpt/gpt-4o");
     expect(ids).toContain("web-chatgpt/gpt-4o-mini");
-    expect(ids).toContain("web-chatgpt/gpt-o3");
-    expect(ids).toContain("web-chatgpt/gpt-o4-mini");
+    expect(ids).toContain("web-chatgpt/gpt-4.1");
+    expect(ids).toContain("web-chatgpt/o3");
+    expect(ids).toContain("web-chatgpt/o4-mini");
     expect(ids).toContain("web-chatgpt/gpt-5");
+    expect(ids).toContain("web-chatgpt/gpt-5-mini");
   });
 
   it("web-chatgpt/* models listed in CLI_MODELS constant", () => {
     const chatgpt = CLI_MODELS.filter(m => m.id.startsWith("web-chatgpt/"));
-    expect(chatgpt).toHaveLength(5);
+    expect(chatgpt).toHaveLength(7);
   });
 });