v2.6.0: API spec updates, free/open-source branding, dependency upgrades

- Add max_completion_tokens normalization (GPT-5.x compat)
- Add developer role → system mapping (OpenAI 2025+ spec)
- Add stream_options.include_usage for streaming responses
- Responses API: text.format and truncation pass-through
- Expand Anthropic interface (thinking, metadata, tool_use)
- Expand Google interface (frequencyPenalty, presencePenalty, safetySettings)
- Update branding to emphasize free, open-source, trustworthy
- Upgrade typescript 5.9.3, typescript-eslint 8.55.0, @types/node 25.2.3

Author: suhaibbinyounis

CHANGELOG.md

@@ -4,6 +4,21 @@ All notable changes to the "github-copilot-api-vscode" extension will be documen
 
 Check [Keep a Changelog](http://keepachangelog.com/) for recommendations on how to structure this file.
 
+## [2.6.0] - 2026-02-13
+
+### Added
+- **OpenAI Chat Completions:** `max_completion_tokens` parameter support (auto-normalized to `max_tokens` for GPT-5.x compatibility)
+- **OpenAI Chat Completions:** `developer` role support (auto-mapped to `system` for 2025+ spec)
+- **OpenAI Chat Completions:** `stream_options.include_usage` — emit usage chunk in streaming responses
+- **OpenAI Responses API:** `text.format` structured output pass-through (was hardcoded to `text`)
+- **OpenAI Responses API:** `truncation` parameter pass-through
+- **Anthropic Messages API:** `thinking`, `metadata` interface fields; `tool_use` stop reason; cache token usage
+- **Google Generative AI:** `frequencyPenalty`, `presencePenalty`, `responseMimeType`, `responseSchema`, `safetySettings` interface
+
+### Changed
+- Updated branding to emphasize free, open-source, and trustworthy nature
+- Upgraded TypeScript to 5.9.3, typescript-eslint to 8.55.0, @types/node to 25.2.3
+
 ## [2.5.1] - 2026-02-07
 
 ### Fixed

README.md

@@ -5,8 +5,8 @@
 <h1 align="center">GitHub Copilot API Gateway</h1>
 
 <p align="center">
-  <strong>Use GitHub Copilot like any other AI API.</strong><br>
-  One VS Code extension. Zero API keys. Works with LangChain, Cursor, Aider, and 50+ tools.
+  <strong>Use GitHub Copilot like any other AI API — completely free & open-source.</strong><br>
+  One VS Code extension. Zero API keys. 100% transparent. Works with LangChain, Cursor, Aider, and 50+ tools.
 </p>
 
 <p align="center">
@@ -67,6 +67,8 @@ response = client.chat.completions.create(
 )
 ```
 
+> 🔒 **Your data stays local.** This extension runs entirely on your machine — no third-party servers, no telemetry, no data collection. The full source code is available on [GitHub](https://github.com/suhaibbinyounis/github-copilot-api-vscode) for you to audit.
+
 ---
 
 ## ✅ What You Can Do
@@ -174,11 +176,17 @@ For advanced options (TLS, IP allowlisting, audit logging), see [CONTRIBUTING.md
 
 ---
 
-## 📝 License
+## 🛡️ Trust & Transparency
+
+This project is **100% free and open-source** under the [MIT License](LICENSE).
 
-**MIT License** — See [LICENSE](LICENSE) for details.
+- 🔍 **Full source code** available on [GitHub](https://github.com/suhaibbinyounis/github-copilot-api-vscode)
+- 🚫 **No telemetry** — zero data collection, no analytics, no tracking
+- 🔒 **Runs locally** — your prompts and API keys never leave your machine
+- ✅ **Community-driven** — contributions welcome via [CONTRIBUTING.md](CONTRIBUTING.md)
+- 🛡️ **Security** — report vulnerabilities via [SECURITY.md](SECURITY.md)
 
-> **Disclaimer:** This extension is an independent project and is not affiliated with GitHub, Microsoft, or OpenAI. It leverages your existing GitHub Copilot subscription. Use responsibly.
+> **Disclaimer:** This extension is an independent, community-built project and is not affiliated with GitHub, Microsoft, or OpenAI. It leverages your existing GitHub Copilot subscription. Use responsibly.
 
 <p align="center">
   <strong>Built with ❤️ by <a href="https://suhaibbinyounis.com">Suhaib Bin Younis</a></strong>

dist/extension.js

@@ -52632,6 +52632,16 @@ var CopilotApiGateway = class {
     }
     if (req.method === "POST" && url2.pathname === "/v1/chat/completions") {
       const body = await this.readJsonBody(req);
+      if (body?.max_completion_tokens && !body?.max_tokens) {
+        body.max_tokens = body.max_completion_tokens;
+      }
+      if (body?.messages && Array.isArray(body.messages)) {
+        for (const msg of body.messages) {
+          if (msg.role === "developer") {
+            msg.role = "system";
+          }
+        }
+      }
       if (body?.stream === true) {
         await this.processStreamingChatCompletion(body, req, res, requestId, requestStart);
       } else {
@@ -52787,6 +52797,13 @@ var CopilotApiGateway = class {
       if (body?.max_completion_tokens && !body?.max_tokens) {
         body.max_tokens = body.max_completion_tokens;
       }
+      if (body?.messages && Array.isArray(body.messages)) {
+        for (const msg of body.messages) {
+          if (msg.role === "developer") {
+            msg.role = "system";
+          }
+        }
+      }
       if (body?.stream === true) {
         await this.processStreamingChatCompletion(body, req, res, requestId, requestStart);
       } else {
@@ -53242,6 +53259,33 @@ data: ${JSON.stringify({ type: "error", error: { type: apiError.code || "api_err
       res.write(`data: ${JSON.stringify(finalChunk)}
 
 `);
+      if (payload?.stream_options?.include_usage) {
+        let tokensIn2 = 0;
+        let tokensOut2 = 0;
+        try {
+          const inputString = lmMessages.map((m) => {
+            return typeof m.content === "string" ? m.content : JSON.stringify(m.content);
+          }).join("\n");
+          tokensIn2 = await lmModel.countTokens(inputString, new vscode4.CancellationTokenSource().token);
+          tokensOut2 = await lmModel.countTokens(totalContent, new vscode4.CancellationTokenSource().token);
+        } catch (_) {
+        }
+        const usageChunk = {
+          id: requestId,
+          object: "chat.completion.chunk",
+          created,
+          model,
+          choices: [],
+          usage: {
+            prompt_tokens: tokensIn2,
+            completion_tokens: tokensOut2,
+            total_tokens: tokensIn2 + tokensOut2
+          }
+        };
+        res.write(`data: ${JSON.stringify(usageChunk)}
+
+`);
+      }
       res.write("data: [DONE]\n\n");
       res.end();
       let tokensIn = 0;
@@ -53400,14 +53444,14 @@ data: ${JSON.stringify({ type: "error", error: { type: apiError.code || "api_err
       store: payload.store ?? true,
       temperature: payload.temperature ?? 1,
       text: {
-        format: {
+        format: payload.text?.format ?? {
           type: "text"
         }
       },
       tool_choice: payload.tool_choice ?? "auto",
       tools: payload.tools ?? [],
       top_p: payload.top_p ?? 1,
-      truncation: "disabled",
+      truncation: payload.truncation ?? "disabled",
       usage: {
         input_tokens: inputTokens,
         input_tokens_details: {