dragon-cli 0.2.0__tar.gz → 0.2.1__tar.gz

{dragon_cli-0.2.0/src/dragon_cli.egg-info → dragon_cli-0.2.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dragon-cli
-Version: 0.2.0
+Version: 0.2.1
 Summary: CLI to chat with local GGUF LLMs, with MCP tool-calling and SQLite conversation history.
 Author: JPEDROPS092
 License-Expression: MIT
@@ -17,14 +17,14 @@ Requires-Dist: mcp
 Requires-Dist: fastapi
 Requires-Dist: starlette
 Requires-Dist: uvicorn
-Requires-Dist: fastapi
+Requires-Dist: jinja2
 Provides-Extra: llm
 Requires-Dist: llama-cpp-python; extra == "llm"
 Dynamic: license-file
 
 # Dragon CLI
 
-![1785346956957](image/README/1785346956957.png)
+![1785346956957](https://raw.githubusercontent.com/JPEDROPS092/dragon-cli-ai-local/main/image/README/1785346956957.png)
 
 [![License: MIT](https://img.shields.io/github/license/JPEDROPS092/dragon-cli-ai-local)](LICENSE)
 [![Python 3.10+](https://img.shields.io/badge/python-3.10%2B-blue)](https://www.python.org/)
@@ -32,38 +32,61 @@ Dynamic: license-file
 [![GitHub stars](https://img.shields.io/github/stars/JPEDROPS092/dragon-cli-ai-local?style=social)](https://github.com/JPEDROPS092/dragon-cli-ai-local/stargazers)
 [![GitHub issues](https://img.shields.io/github/issues/JPEDROPS092/dragon-cli-ai-local)](https://github.com/JPEDROPS092/dragon-cli-ai-local/issues)
 
-Chat with local GGUF LLMs (100% CPU, via `llama-cpp-python`), with optional
-MCP (Model Context Protocol) tool-calling and a SQLite-backed conversation
-history.
+[![PyPI](https://img.shields.io/pypi/v/dragon-cli)](https://pypi.org/project/dragon-cli/)
+
+Chat with local GGUF LLMs on CPU or GPU (via `llama-cpp-python`), with
+MCP (Model Context Protocol) tool-calling, a SQLite-backed conversation
+history, and an OpenAI-compatible HTTP server you can plug into agent
+frameworks (OpenAI SDK, OpenAI Agents SDK, LangChain, LlamaIndex, CrewAI,
+Continue, Aider, Open WebUI...).
+
+📖 **Full docs, guides and agent-framework examples:**
+https://jpedrops092.github.io/dragon-cli-ai-local/
 
 ## Install
 
+Requires Python 3.10+ on Windows, Linux or macOS.
+
 ```bash
-pip install -e .
-dragon-cli-install-llm
+pip install dragon-cli
+dragon-cli-install-llm   # installs the right llama-cpp-python build for your CPU/GPU
+dragon-cli               # interactive menu
 ```
 
-This exposes a `dragon-cli` command. `llama-cpp-python` is intentionally kept
-out of the core dependencies, since it ships as C++ source on PyPI for many
-platforms — a plain `pip install` would try to compile it locally and fail
-without a C/C++ toolchain (Visual Studio Build Tools on Windows, gcc on
-Linux, Xcode command line tools on macOS).
+Upgrade with `pip install -U dragon-cli`. A virtual environment is
+recommended (`python -m venv .venv` and activate it first).
 
-`dragon-cli-install-llm` detects your OS/architecture (Windows, Linux, or
-macOS) and installs a prebuilt CPU wheel instead, so a compiler usually isn't
-needed at all. If no prebuilt wheel is available for your exact
+`llama-cpp-python` is intentionally kept out of the core dependencies, since
+it ships as C++ source on PyPI for many platforms, so a plain `pip install`
+would try to compile it locally and fail without a C/C++ toolchain (Visual
+Studio Build Tools on Windows, gcc on Linux, Xcode command line tools on
+macOS).
+
+`dragon-cli-install-llm` detects your OS/architecture and GPU (NVIDIA CUDA,
+AMD ROCm, Apple Metal) and installs a prebuilt wheel instead, so a compiler
+usually isn't needed at all. If no prebuilt wheel is available for your exact
 platform/Python combination, it prints the toolchain to install and the
 fallback command to run.
 
 If you already have a working C/C++ toolchain and would rather build from
-source yourself, skip the installer and use the extra instead:
-`pip install -e ".[llm]"`.
+source yourself: `pip install "dragon-cli[llm]"`.
+
+Models (`./models`), `mcp_servers.json` and `dragon_cli.db` are created in
+the folder you run `dragon-cli` from.
 
-You can also run it without installing:
+### Other ways to install
 
 ```bash
+# latest development version from GitHub
+pip install git+https://github.com/JPEDROPS092/dragon-cli-ai-local
+
+# from a local clone (editable)
+git clone https://github.com/JPEDROPS092/dragon-cli-ai-local
+cd dragon-cli-ai-local
+pip install -e .
+
+# or run without installing
 python dragon_cli.py
-# or
 python -m dragon_cli
 ```
 
@@ -80,6 +103,12 @@ dragon-cli --db path/to/other.db # use a different conversation database
 dragon-cli --serve               # serve the model over HTTP instead of chatting
 dragon-cli --serve --port 8080   # custom port (default: 8000)
 dragon-cli --serve --host 0.0.0.0  # expose beyond localhost (e.g. to your LAN) — opt-in only
+dragon-cli --profile             # show detected CPU/RAM/GPU and exit
+dragon-cli --recommend           # rank catalog models for your hardware
+dragon-cli --auto                # pick the best model for your hardware and chat
+dragon-cli --priority speed      # speed | quality | balanced | memory (for --auto/--recommend)
+dragon-cli --gpu / --cpu         # force full GPU offload, or CPU only
+dragon-cli --ngl 20              # offload exactly 20 layers to the GPU
 ```
 
 ## HTTP server (`--serve`)

{dragon_cli-0.2.0 → dragon_cli-0.2.1}/README.md

@@ -1,6 +1,6 @@
 # Dragon CLI
 
-![1785346956957](image/README/1785346956957.png)
+![1785346956957](https://raw.githubusercontent.com/JPEDROPS092/dragon-cli-ai-local/main/image/README/1785346956957.png)
 
 [![License: MIT](https://img.shields.io/github/license/JPEDROPS092/dragon-cli-ai-local)](LICENSE)
 [![Python 3.10+](https://img.shields.io/badge/python-3.10%2B-blue)](https://www.python.org/)
@@ -8,38 +8,61 @@
 [![GitHub stars](https://img.shields.io/github/stars/JPEDROPS092/dragon-cli-ai-local?style=social)](https://github.com/JPEDROPS092/dragon-cli-ai-local/stargazers)
 [![GitHub issues](https://img.shields.io/github/issues/JPEDROPS092/dragon-cli-ai-local)](https://github.com/JPEDROPS092/dragon-cli-ai-local/issues)
 
-Chat with local GGUF LLMs (100% CPU, via `llama-cpp-python`), with optional
-MCP (Model Context Protocol) tool-calling and a SQLite-backed conversation
-history.
+[![PyPI](https://img.shields.io/pypi/v/dragon-cli)](https://pypi.org/project/dragon-cli/)
+
+Chat with local GGUF LLMs on CPU or GPU (via `llama-cpp-python`), with
+MCP (Model Context Protocol) tool-calling, a SQLite-backed conversation
+history, and an OpenAI-compatible HTTP server you can plug into agent
+frameworks (OpenAI SDK, OpenAI Agents SDK, LangChain, LlamaIndex, CrewAI,
+Continue, Aider, Open WebUI...).
+
+📖 **Full docs, guides and agent-framework examples:**
+https://jpedrops092.github.io/dragon-cli-ai-local/
 
 ## Install
 
+Requires Python 3.10+ on Windows, Linux or macOS.
+
 ```bash
-pip install -e .
-dragon-cli-install-llm
+pip install dragon-cli
+dragon-cli-install-llm   # installs the right llama-cpp-python build for your CPU/GPU
+dragon-cli               # interactive menu
 ```
 
-This exposes a `dragon-cli` command. `llama-cpp-python` is intentionally kept
-out of the core dependencies, since it ships as C++ source on PyPI for many
-platforms — a plain `pip install` would try to compile it locally and fail
-without a C/C++ toolchain (Visual Studio Build Tools on Windows, gcc on
-Linux, Xcode command line tools on macOS).
+Upgrade with `pip install -U dragon-cli`. A virtual environment is
+recommended (`python -m venv .venv` and activate it first).
+
+`llama-cpp-python` is intentionally kept out of the core dependencies, since
+it ships as C++ source on PyPI for many platforms, so a plain `pip install`
+would try to compile it locally and fail without a C/C++ toolchain (Visual
+Studio Build Tools on Windows, gcc on Linux, Xcode command line tools on
+macOS).
 
-`dragon-cli-install-llm` detects your OS/architecture (Windows, Linux, or
-macOS) and installs a prebuilt CPU wheel instead, so a compiler usually isn't
-needed at all. If no prebuilt wheel is available for your exact
+`dragon-cli-install-llm` detects your OS/architecture and GPU (NVIDIA CUDA,
+AMD ROCm, Apple Metal) and installs a prebuilt wheel instead, so a compiler
+usually isn't needed at all. If no prebuilt wheel is available for your exact
 platform/Python combination, it prints the toolchain to install and the
 fallback command to run.
 
 If you already have a working C/C++ toolchain and would rather build from
-source yourself, skip the installer and use the extra instead:
-`pip install -e ".[llm]"`.
+source yourself: `pip install "dragon-cli[llm]"`.
 
-You can also run it without installing:
+Models (`./models`), `mcp_servers.json` and `dragon_cli.db` are created in
+the folder you run `dragon-cli` from.
+
+### Other ways to install
 
 ```bash
+# latest development version from GitHub
+pip install git+https://github.com/JPEDROPS092/dragon-cli-ai-local
+
+# from a local clone (editable)
+git clone https://github.com/JPEDROPS092/dragon-cli-ai-local
+cd dragon-cli-ai-local
+pip install -e .
+
+# or run without installing
 python dragon_cli.py
-# or
 python -m dragon_cli
 ```
 
@@ -56,6 +79,12 @@ dragon-cli --db path/to/other.db # use a different conversation database
 dragon-cli --serve               # serve the model over HTTP instead of chatting
 dragon-cli --serve --port 8080   # custom port (default: 8000)
 dragon-cli --serve --host 0.0.0.0  # expose beyond localhost (e.g. to your LAN) — opt-in only
+dragon-cli --profile             # show detected CPU/RAM/GPU and exit
+dragon-cli --recommend           # rank catalog models for your hardware
+dragon-cli --auto                # pick the best model for your hardware and chat
+dragon-cli --priority speed      # speed | quality | balanced | memory (for --auto/--recommend)
+dragon-cli --gpu / --cpu         # force full GPU offload, or CPU only
+dragon-cli --ngl 20              # offload exactly 20 layers to the GPU
 ```
 
 ## HTTP server (`--serve`)

{dragon_cli-0.2.0 → dragon_cli-0.2.1}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "dragon-cli"
-version = "0.2.0"
+version = "0.2.1"
 description = "CLI to chat with local GGUF LLMs, with MCP tool-calling and SQLite conversation history."
 readme = "README.md"
 requires-python = ">=3.10"
@@ -18,7 +18,7 @@ dependencies = [
     "fastapi",
     "starlette",
     "uvicorn",
-    "fastapi"
+    "jinja2",
 ]
 
 [project.optional-dependencies]
@@ -40,3 +40,6 @@ dragon-cli-install-llm = "dragon_cli.setup_llm:main"
 
 [tool.setuptools.packages.find]
 where = ["src"]
+
+[tool.setuptools.package-data]
+dragon_cli = ["templates/*", "static/*"]

{dragon_cli-0.2.0 → dragon_cli-0.2.1}/src/dragon_cli/__init__.py

@@ -1,4 +1,4 @@
-__version__ = "0.2.0"
+__version__ = "0.2.1"
 
 from ._cuda import preload_cuda_runtime
 

dragon_cli-0.2.1/src/dragon_cli/static/style.css

@@ -0,0 +1,352 @@
+:root {
+  color-scheme: dark;
+  font-family: Inter, ui-sans-serif, system-ui, -apple-system, BlinkMacSystemFont, "Segoe UI", sans-serif;
+  background: #09101a;
+  color: #e7edf4;
+}
+
+* {
+  box-sizing: border-box;
+}
+
+body {
+  margin: 0;
+  min-height: 100vh;
+  padding: 0;
+  background: radial-gradient(circle at top, rgba(93, 173, 249, 0.15), transparent 35%),
+              linear-gradient(180deg, #08111c 0%, #0b141e 100%);
+}
+
+button,
+input,
+textarea {
+  font: inherit;
+}
+
+.app-shell {
+  display: grid;
+  grid-template-columns: 280px 1fr;
+  min-height: 100vh;
+  gap: 1rem;
+  padding: 1.25rem;
+}
+
+.sidebar {
+  background: rgba(14, 24, 38, 0.95);
+  border: 1px solid rgba(255, 255, 255, 0.06);
+  border-radius: 24px;
+  padding: 1.5rem;
+  display: flex;
+  flex-direction: column;
+  gap: 1.5rem;
+}
+
+.brand {
+  display: flex;
+  align-items: center;
+  gap: 1rem;
+}
+
+.brand-icon {
+  width: 54px;
+  height: 54px;
+  object-fit: contain;
+  border-radius: 18px;
+  background: linear-gradient(135deg, #2c5a7b, #0f2334);
+  border: 1px solid rgba(255, 255, 255, 0.08);
+  padding: 0.7rem;
+}
+
+.brand h1 {
+  margin: 0;
+  font-size: 1.3rem;
+}
+
+.brand p {
+  margin: 0.25rem 0 0;
+  color: #8fabc6;
+}
+
+.model-card,
+.instructions {
+  background: rgba(255, 255, 255, 0.03);
+  border: 1px solid rgba(255, 255, 255, 0.08);
+  border-radius: 18px;
+  padding: 1rem;
+}
+
+.model-card strong {
+  display: block;
+  margin-bottom: 0.5rem;
+  color: #90d7ff;
+}
+
+.chat-panel {
+  display: flex;
+  flex-direction: column;
+  background: rgba(6, 14, 24, 0.95);
+  border: 1px solid rgba(255, 255, 255, 0.06);
+  border-radius: 28px;
+  padding: 1.25rem;
+  min-height: calc(100vh - 2.5rem);
+}
+
+.chat-header {
+  display: flex;
+  justify-content: space-between;
+  align-items: flex-start;
+  gap: 1rem;
+  margin-bottom: 1rem;
+}
+
+.chat-header h2 {
+  margin: 0;
+  font-size: 1.35rem;
+}
+
+.chat-header p {
+  margin: 0.35rem 0 0;
+  color: #8da3c4;
+}
+
+.messages {
+  flex: 1;
+  display: grid;
+  gap: 0.85rem;
+  overflow-y: auto;
+  padding-right: 0.25rem;
+}
+
+.message {
+  padding: 1rem 1.1rem;
+  border-radius: 18px;
+  line-height: 1.6;
+  white-space: pre-wrap;
+  word-break: break-word;
+}
+
+.message-user {
+  background: rgba(68, 119, 218, 0.16);
+  border: 1px solid rgba(68, 119, 218, 0.25);
+  align-self: end;
+}
+
+.message-assistant {
+  background: rgba(98, 234, 194, 0.08);
+  border: 1px solid rgba(98, 234, 194, 0.14);
+}
+
+.message-role {
+  margin-bottom: 0.5rem;
+  font-size: 0.82rem;
+  letter-spacing: 0.04em;
+  text-transform: uppercase;
+  color: #72d6c9;
+}
+
+.composer {
+  margin-top: 1rem;
+  display: grid;
+  gap: 0.9rem;
+}
+
+textarea {
+  width: 100%;
+  min-height: 120px;
+  resize: vertical;
+  border-radius: 20px;
+  border: 1px solid rgba(255, 255, 255, 0.08);
+  background: rgba(10, 19, 32, 0.92);
+  color: #eef4fb;
+  padding: 1rem 1rem;
+}
+
+textarea:focus {
+  outline: none;
+  border-color: #5dd9f9;
+  box-shadow: 0 0 0 3px rgba(93, 217, 255, 0.12);
+}
+
+.composer-actions {
+  display: flex;
+  justify-content: space-between;
+  align-items: center;
+  gap: 1rem;
+  flex-wrap: wrap;
+}
+
+button[type="submit"] {
+  border: none;
+  cursor: pointer;
+  border-radius: 999px;
+  padding: 0.95rem 1.55rem;
+  background: linear-gradient(135deg, #2b9bff, #4fc3ff);
+  color: #081116;
+  font-weight: 700;
+}
+
+button[type="submit"]:hover {
+  opacity: 0.95;
+}
+
+label {
+  display: inline-flex;
+  align-items: center;
+  gap: 0.5rem;
+  color: #b7c8de;
+  font-size: 0.95rem;
+}
+
+input[type="checkbox"] {
+  width: 16px;
+  height: 16px;
+}
+
+@media (max-width: 980px) {
+  .app-shell {
+    grid-template-columns: 1fr;
+  }
+  .chat-panel {
+    min-height: auto;
+  }
+}
+
+/* ---------- Welcome page ---------- */
+.welcome-shell {
+  max-width: 920px;
+  margin: 0 auto;
+  padding: 3rem 1.25rem 2rem;
+  display: flex;
+  flex-direction: column;
+  gap: 2rem;
+}
+
+.welcome-hero {
+  text-align: center;
+  background: rgba(14, 24, 38, 0.95);
+  border: 1px solid rgba(255, 255, 255, 0.06);
+  border-radius: 28px;
+  padding: 2.5rem 2rem;
+}
+
+.welcome-icon {
+  width: 80px;
+  height: 80px;
+  object-fit: contain;
+  border-radius: 22px;
+  background: linear-gradient(135deg, #2c5a7b, #0f2334);
+  border: 1px solid rgba(255, 255, 255, 0.08);
+  padding: 0.9rem;
+}
+
+.welcome-hero h1 {
+  margin: 1.25rem 0 0.5rem;
+  font-size: 2rem;
+}
+
+.welcome-sub {
+  margin: 0 auto 1.5rem;
+  max-width: 560px;
+  color: #8fabc6;
+  line-height: 1.6;
+}
+
+.cta {
+  display: inline-block;
+  text-decoration: none;
+  border-radius: 999px;
+  padding: 0.85rem 1.6rem;
+  background: linear-gradient(135deg, #2b9bff, #4fc3ff);
+  color: #081116;
+  font-weight: 700;
+}
+
+.cta:hover {
+  opacity: 0.95;
+}
+
+.welcome-grid {
+  display: grid;
+  grid-template-columns: repeat(auto-fit, minmax(280px, 1fr));
+  gap: 1.25rem;
+}
+
+.card {
+  background: rgba(255, 255, 255, 0.03);
+  border: 1px solid rgba(255, 255, 255, 0.08);
+  border-radius: 20px;
+  padding: 1.5rem;
+}
+
+.card h2 {
+  margin: 0 0 0.85rem;
+  font-size: 1.1rem;
+  color: #90d7ff;
+}
+
+.card p {
+  margin: 0 0 0.75rem;
+  line-height: 1.6;
+  color: #b7c8de;
+}
+
+.endpoint-list {
+  margin: 0;
+  padding: 0;
+  list-style: none;
+  display: grid;
+  gap: 0.6rem;
+}
+
+.endpoint-list li {
+  color: #b7c8de;
+  line-height: 1.5;
+}
+
+.code-block {
+  margin: 0.25rem 0 0;
+  padding: 1rem;
+  background: rgba(6, 14, 24, 0.95);
+  border: 1px solid rgba(255, 255, 255, 0.06);
+  border-radius: 14px;
+  color: #d6e7f5;
+  font-family: ui-monospace, "SFMono-Regular", Menlo, Consolas, monospace;
+  font-size: 0.85rem;
+  line-height: 1.5;
+  overflow-x: auto;
+  white-space: pre;
+}
+
+.welcome-footer {
+  display: flex;
+  flex-wrap: wrap;
+  gap: 0.6rem;
+  justify-content: center;
+  align-items: center;
+  color: #8fabc6;
+  font-size: 0.9rem;
+  padding-top: 0.5rem;
+}
+
+.welcome-footer a {
+  color: #90d7ff;
+  text-decoration: none;
+}
+
+.welcome-footer a:hover {
+  text-decoration: underline;
+}
+
+.instructions a,
+.welcome-footer a {
+  color: #90d7ff;
+}
+
+code {
+  background: rgba(255, 255, 255, 0.06);
+  border-radius: 6px;
+  padding: 0.1rem 0.35rem;
+  font-family: ui-monospace, "SFMono-Regular", Menlo, Consolas, monospace;
+  font-size: 0.85em;
+}
+

dragon_cli-0.2.1/src/dragon_cli/templates/chat.html

@@ -0,0 +1,160 @@
+<!doctype html>
+<html lang="en">
+<head>
+  <meta charset="utf-8">
+  <meta name="viewport" content="width=device-width, initial-scale=1">
+  <title>Dragon CLI Chat UI</title>
+  <link rel="icon" type="image/svg+xml" href="/favicon.ico">
+  <link rel="stylesheet" href="/static/style.css">
+</head>
+<body>
+  <div class="app-shell">
+    <aside class="sidebar">
+      <div class="brand">
+        <img src="/static/dragon-icon.png" alt="Dragon CLI" class="brand-icon">
+        <div>
+          <h1>Dragon CLI</h1>
+          <p>Chat with your local model</p>
+        </div>
+      </div>
+      <div class="model-card">
+        <strong>Model:</strong>
+        <span>{{ model_name }}</span>
+      </div>
+      <div class="instructions">
+        <p>Type your message below and send it to chat with the local model.</p>
+        <p><a href="{{ base_url }}/">Back to home</a></p>
+      </div>
+    </aside>
+
+    <main class="chat-panel">
+      <header class="chat-header">
+        <div>
+          <h2>Dragon CLI Chat</h2>
+          <p>Servidor: {{ base_url }}</p>
+        </div>
+      </header>
+
+      <section id="messages" class="messages" aria-live="polite"></section>
+
+      <form id="chat-form" class="composer">
+        <textarea id="prompt" placeholder="Type your question here..." rows="3" required></textarea>
+        <div class="composer-actions">
+          <button type="submit">Send</button>
+          <label>
+            <input type="checkbox" id="stream" checked>
+            Streaming
+          </label>
+        </div>
+      </form>
+    </main>
+  </div>
+
+  <script>
+    const baseUrl = "{{ base_url }}";
+    const modelName = "{{ model_name }}";
+
+    const messagesEl = document.getElementById("messages");
+    const form = document.getElementById("chat-form");
+    const promptInput = document.getElementById("prompt");
+    const streamCheckbox = document.getElementById("stream");
+
+    function addMessage(role, text) {
+      const messageElement = document.createElement("div");
+      messageElement.className = `message message-${role}`;
+      messageElement.innerHTML = `<div class="message-role">${role}</div><div class="message-text"></div>`;
+      messageElement.querySelector(".message-text").textContent = text;
+      messagesEl.appendChild(messageElement);
+      messagesEl.scrollTop = messagesEl.scrollHeight;
+      return messageElement;
+    }
+
+    function setAssistantText(el, text) {
+      el.querySelector(".message-text").textContent = text;
+      messagesEl.scrollTop = messagesEl.scrollHeight;
+    }
+
+    form.addEventListener("submit", async (event) => {
+      event.preventDefault();
+      const prompt = promptInput.value.trim();
+      if (!prompt) return;
+
+      const userMessage = addMessage("user", prompt);
+      promptInput.value = "";
+      promptInput.focus();
+
+      const assistantMessage = addMessage("assistant", "Waiting for response...");
+      const stream = streamCheckbox.checked;
+
+      try {
+        if (stream) {
+          const res = await fetch(`${baseUrl}/v1/chat/completions`, {
+            method: "POST",
+            headers: { "Content-Type": "application/json" },
+            body: JSON.stringify({
+              model: modelName,
+              messages: [
+                { role: "user", content: prompt }
+              ],
+              stream: true,
+            }),
+          });
+
+          if (!res.ok) {
+            throw new Error(`Error ${res.status}`);
+          }
+
+          const reader = res.body.getReader();
+          const decoder = new TextDecoder("utf-8");
+          let assistantText = "";
+
+          while (true) {
+            const { done, value } = await reader.read();
+            if (done) break;
+            const chunk = decoder.decode(value, { stream: true });
+            for (const line of chunk.split("\n")) {
+              if (!line.trim().startsWith("data:")) continue;
+              const payload = line.replace(/^data:\s*/, "");
+              if (payload === "[DONE]") {
+                return;
+              }
+              try {
+                const event = JSON.parse(payload);
+                if (event.choices?.[0]?.delta?.content) {
+                  assistantText += event.choices[0].delta.content;
+                  setAssistantText(assistantMessage, assistantText);
+                }
+              } catch (error) {
+                console.warn("Failed to parse SSE chunk", error);
+              }
+            }
+          }
+        } else {
+          const res = await fetch(`${baseUrl}/v1/chat/completions`, {
+            method: "POST",
+            headers: { "Content-Type": "application/json" },
+            body: JSON.stringify({
+              model: modelName,
+              messages: [
+                { role: "user", content: prompt }
+              ],
+              stream: false,
+            }),
+          });
+
+          if (!res.ok) {
+            const details = await res.text();
+            throw new Error(details || `Error ${res.status}`);
+          }
+
+          const payload = await res.json();
+          const content = payload.choices?.[0]?.message?.content || "No response.";
+          setAssistantText(assistantMessage, content);
+        }
+      } catch (error) {
+        setAssistantText(assistantMessage, `Error: ${error.message}`);
+      }
+    });
+  </script>
+</body>
+</html>

dragon_cli-0.2.1/src/dragon_cli/templates/index.html

@@ -0,0 +1,82 @@
+<!doctype html>
+<html lang="en">
+<head>
+  <meta charset="utf-8">
+  <meta name="viewport" content="width=device-width, initial-scale=1">
+  <title>Dragon CLI — Local Model Server</title>
+  <link rel="icon" type="image/svg+xml" href="/favicon.ico">
+  <link rel="stylesheet" href="/static/style.css">
+</head>
+<body>
+  <div class="welcome-shell">
+    <header class="welcome-hero">
+      <img src="/static/dragon-icon.png" alt="Dragon CLI" class="welcome-icon">
+      <h1>Welcome to Dragon CLI</h1>
+      <p class="welcome-sub">
+        Local OpenAI-compatible server running the
+        <strong>{{ model_name }}</strong> model.
+      </p>
+      <a class="cta" href="{{ base_url }}/chat">Open chat &rarr;</a>
+    </header>
+
+    <section class="welcome-grid">
+      <article class="card">
+        <h2>Main endpoints</h2>
+        <ul class="endpoint-list">
+          <li><code>GET&nbsp;&nbsp;/v1/models</code> — list the loaded model</li>
+          <li><code>POST&nbsp;/v1/chat/completions</code> — chat completion (stream or not)</li>
+          <li><code>GET&nbsp;&nbsp;/docs</code> — interactive docs (Swagger UI)</li>
+          <li><code>GET&nbsp;&nbsp;/chat</code> — web chat interface</li>
+        </ul>
+      </article>
+
+      <article class="card">
+        <h2>How to use</h2>
+        <p>
+          Point any OpenAI client at
+          <code>{{ base_url }}/v1</code> using <em>any key</em> (authentication is ignored).
+          The full base URL is:
+        </p>
+        <pre class="code-block">{{ base_url }}/v1</pre>
+      </article>
+
+      <article class="card">
+        <h2>cURL example</h2>
+        <pre class="code-block">curl -X POST {{ base_url }}/v1/chat/completions \
+  -H "Content-Type: application/json" \
+  -d '{
+    "model": "{{ model_name }}",
+    "messages": [
+      {"role": "user", "content": "Hello!"}
+    ],
+    "stream": false
+  }'</pre>
+      </article>
+
+      <article class="card">
+        <h2>Python example</h2>
+        <pre class="code-block">from openai import OpenAI
+
+client = OpenAI(
+    base_url="{{ base_url }}/v1",
+    api_key="ignored",
+)
+
+resp = client.chat.completions.create(
+    model="{{ model_name }}",
+    messages=[{"role": "user", "content": "Hello!"}],
+)
+print(resp.choices[0].message.content)</pre>
+      </article>
+    </section>
+
+    <footer class="welcome-footer">
+      <span>Active model: <strong>{{ model_name }}</strong></span>
+      <span>&middot;</span>
+      <a href="{{ base_url }}/docs">Swagger UI</a>
+      <span>&middot;</span>
+      <a href="{{ base_url }}/chat">Chat</a>
+    </footer>
+  </div>
+</body>
+</html>

{dragon_cli-0.2.0 → dragon_cli-0.2.1/src/dragon_cli.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dragon-cli
-Version: 0.2.0
+Version: 0.2.1
 Summary: CLI to chat with local GGUF LLMs, with MCP tool-calling and SQLite conversation history.
 Author: JPEDROPS092
 License-Expression: MIT
@@ -17,14 +17,14 @@ Requires-Dist: mcp
 Requires-Dist: fastapi
 Requires-Dist: starlette
 Requires-Dist: uvicorn
-Requires-Dist: fastapi
+Requires-Dist: jinja2
 Provides-Extra: llm
 Requires-Dist: llama-cpp-python; extra == "llm"
 Dynamic: license-file
 
 # Dragon CLI
 
-![1785346956957](image/README/1785346956957.png)
+![1785346956957](https://raw.githubusercontent.com/JPEDROPS092/dragon-cli-ai-local/main/image/README/1785346956957.png)
 
 [![License: MIT](https://img.shields.io/github/license/JPEDROPS092/dragon-cli-ai-local)](LICENSE)
 [![Python 3.10+](https://img.shields.io/badge/python-3.10%2B-blue)](https://www.python.org/)
@@ -32,38 +32,61 @@ Dynamic: license-file
 [![GitHub stars](https://img.shields.io/github/stars/JPEDROPS092/dragon-cli-ai-local?style=social)](https://github.com/JPEDROPS092/dragon-cli-ai-local/stargazers)
 [![GitHub issues](https://img.shields.io/github/issues/JPEDROPS092/dragon-cli-ai-local)](https://github.com/JPEDROPS092/dragon-cli-ai-local/issues)
 
-Chat with local GGUF LLMs (100% CPU, via `llama-cpp-python`), with optional
-MCP (Model Context Protocol) tool-calling and a SQLite-backed conversation
-history.
+[![PyPI](https://img.shields.io/pypi/v/dragon-cli)](https://pypi.org/project/dragon-cli/)
+
+Chat with local GGUF LLMs on CPU or GPU (via `llama-cpp-python`), with
+MCP (Model Context Protocol) tool-calling, a SQLite-backed conversation
+history, and an OpenAI-compatible HTTP server you can plug into agent
+frameworks (OpenAI SDK, OpenAI Agents SDK, LangChain, LlamaIndex, CrewAI,
+Continue, Aider, Open WebUI...).
+
+📖 **Full docs, guides and agent-framework examples:**
+https://jpedrops092.github.io/dragon-cli-ai-local/
 
 ## Install
 
+Requires Python 3.10+ on Windows, Linux or macOS.
+
 ```bash
-pip install -e .
-dragon-cli-install-llm
+pip install dragon-cli
+dragon-cli-install-llm   # installs the right llama-cpp-python build for your CPU/GPU
+dragon-cli               # interactive menu
 ```
 
-This exposes a `dragon-cli` command. `llama-cpp-python` is intentionally kept
-out of the core dependencies, since it ships as C++ source on PyPI for many
-platforms — a plain `pip install` would try to compile it locally and fail
-without a C/C++ toolchain (Visual Studio Build Tools on Windows, gcc on
-Linux, Xcode command line tools on macOS).
+Upgrade with `pip install -U dragon-cli`. A virtual environment is
+recommended (`python -m venv .venv` and activate it first).
 
-`dragon-cli-install-llm` detects your OS/architecture (Windows, Linux, or
-macOS) and installs a prebuilt CPU wheel instead, so a compiler usually isn't
-needed at all. If no prebuilt wheel is available for your exact
+`llama-cpp-python` is intentionally kept out of the core dependencies, since
+it ships as C++ source on PyPI for many platforms, so a plain `pip install`
+would try to compile it locally and fail without a C/C++ toolchain (Visual
+Studio Build Tools on Windows, gcc on Linux, Xcode command line tools on
+macOS).
+
+`dragon-cli-install-llm` detects your OS/architecture and GPU (NVIDIA CUDA,
+AMD ROCm, Apple Metal) and installs a prebuilt wheel instead, so a compiler
+usually isn't needed at all. If no prebuilt wheel is available for your exact
 platform/Python combination, it prints the toolchain to install and the
 fallback command to run.
 
 If you already have a working C/C++ toolchain and would rather build from
-source yourself, skip the installer and use the extra instead:
-`pip install -e ".[llm]"`.
+source yourself: `pip install "dragon-cli[llm]"`.
+
+Models (`./models`), `mcp_servers.json` and `dragon_cli.db` are created in
+the folder you run `dragon-cli` from.
 
-You can also run it without installing:
+### Other ways to install
 
 ```bash
+# latest development version from GitHub
+pip install git+https://github.com/JPEDROPS092/dragon-cli-ai-local
+
+# from a local clone (editable)
+git clone https://github.com/JPEDROPS092/dragon-cli-ai-local
+cd dragon-cli-ai-local
+pip install -e .
+
+# or run without installing
 python dragon_cli.py
-# or
 python -m dragon_cli
 ```
 
@@ -80,6 +103,12 @@ dragon-cli --db path/to/other.db # use a different conversation database
 dragon-cli --serve               # serve the model over HTTP instead of chatting
 dragon-cli --serve --port 8080   # custom port (default: 8000)
 dragon-cli --serve --host 0.0.0.0  # expose beyond localhost (e.g. to your LAN) — opt-in only
+dragon-cli --profile             # show detected CPU/RAM/GPU and exit
+dragon-cli --recommend           # rank catalog models for your hardware
+dragon-cli --auto                # pick the best model for your hardware and chat
+dragon-cli --priority speed      # speed | quality | balanced | memory (for --auto/--recommend)
+dragon-cli --gpu / --cpu         # force full GPU offload, or CPU only
+dragon-cli --ngl 20              # offload exactly 20 layers to the GPU
 ```
 
 ## HTTP server (`--serve`)

{dragon_cli-0.2.0 → dragon_cli-0.2.1}/src/dragon_cli.egg-info/SOURCES.txt

@@ -32,6 +32,10 @@ src/dragon_cli/mcp/tool_call.py
 src/dragon_cli/models/__init__.py
 src/dragon_cli/models/catalog.py
 src/dragon_cli/models/loader.py
+src/dragon_cli/static/dragon-icon.png
+src/dragon_cli/static/style.css
 src/dragon_cli/storage/__init__.py
 src/dragon_cli/storage/conversations.py
-src/dragon_cli/storage/database.py
+src/dragon_cli/storage/database.py
+src/dragon_cli/templates/chat.html
+src/dragon_cli/templates/index.html

{dragon_cli-0.2.0 → dragon_cli-0.2.1}/src/dragon_cli.egg-info/requires.txt

@@ -5,7 +5,7 @@ mcp
 fastapi
 starlette
 uvicorn
-fastapi
+jinja2
 
 [llm]
 llama-cpp-python