amv-train 2.1.0 → 3.0.0

package/NOTICE

@@ -1,8 +1,8 @@
-termux-train
-Copyright 2026 termux-train contributors
+amv-train
+Copyright 2026 amv-train contributors
 
-This product includes software developed by the termux-train project
-(https://github.com/uno-km/termux-train) and AMEVA Open-Source Foundation.
+This product includes software developed by the amv-train project
+(https://github.com/uno-km/amv-train) and AMEVA Open-Source Foundation.
 
 Licensed under the Apache License, Version 2.0 (the "License");
 you may not use this file except in compliance with the License.

package/README.md

@@ -24,7 +24,7 @@
 
 ## Executive Summary & Architecture Overview
 
-**Termux-Train v2.0.0** is an industrial-grade, zero-dependency deep learning framework engineered specifically for constrained ARM64 Bionic environments (Android Termux, mobile Linux, embedded edge devices). By combining a C-vectorized Directed Acyclic Graph (DAG) Autograd runtime with mobile hardware acceleration (Vulkan Compute, OpenCL, ARM64 NEON), Termux-Train enables full backward gradient propagation, parameter-efficient fine-tuning (LoRA / DoRA), and reinforcement learning directly on mobile devices.
+**AMV-Train v2.0.0** is an industrial-grade, zero-dependency deep learning framework engineered specifically for constrained ARM64 Bionic environments (Android Termux, mobile Linux, embedded edge devices). By combining a C-vectorized Directed Acyclic Graph (DAG) Autograd runtime with mobile hardware acceleration (Vulkan Compute, OpenCL, ARM64 NEON), AMV-Train enables full backward gradient propagation, parameter-efficient fine-tuning (LoRA / DoRA), and reinforcement learning directly on mobile devices.
 
 ```
 +---------------------------------------------------------------------------------------------------+
@@ -50,17 +50,17 @@
                                                   v
 +---------------------------------------------------------------------------------------------------+
 |                             Target-Aware Runtime Export & Compatibility                           |
-|    ComfyUI / Diffusers   |   llama.cpp / GGUF   |   termux-vision   |   termux-stt   |   BitNet  |
+|    ComfyUI / Diffusers   |   llama.cpp / GGUF   |   amv-vision   |   amv-stt   |   BitNet  |
 +---------------------------------------------------------------------------------------------------+
 ```
 
 ### Core Architectural Pillars
 
-1. **Zero External Heavy Dependencies**: Executes autonomously without PyTorch, LibTorch, CMake, or heavy BLAS runtimes. Pure Python DAG autograd paired with optional C-accelerator ABI (`libtermux_train_accel.so`) or Vulkan compute shaders.
+1. **Zero External Heavy Dependencies**: Executes autonomously without PyTorch, LibTorch, CMake, or heavy BLAS runtimes. Pure Python DAG autograd paired with optional C-accelerator ABI (`libamv_train_accel.so`) or Vulkan compute shaders.
 2. **6-Modality Target-Aware Adapters**: Generates standard SafeTensors weights and companion `_config.json` manifests natively recognized by major desktop and mobile inference runtimes without conversion tooling.
 3. **Single-Device GPU Slicing**: Bypasses mobile Unified Memory Architecture (UMA) spikes and watchdog timeouts on ARM Mali and Qualcomm Adreno GPUs through granular vocabulary chunking and layer streaming.
 4. **44GB Disaggregated Cluster Virtual RAM Pooling**: Binds physical memory across heterogeneous mobile nodes (e.g., Galaxy S25, S21, S20, A53, A35) into an aggregated pipeline training session protected by a 300MB memory Guard-Band against the Android Low Memory Killer (LMK).
-5. **100% Dual-Engine Parity**: Complete feature and interface symmetry across Python CLI (`amv-train`), Node.js Global CLI (`amv-train`), Python Library (`import termux_train`), and Node.js SDK (`import { ... } from 'amv-train'`).
+5. **100% Dual-Engine Parity**: Complete feature and interface symmetry across Python CLI (`amv-train`), Node.js Global CLI (`amv-train`), Python Library (`import amv_train`), and Node.js SDK (`import { ... } from 'amv-train'`).
 
 ---
 
@@ -104,7 +104,7 @@ amv-train diffusion-train \
 ```
 
 ### 2. Vision Multimodal VLM LoRA Training (`vision-train`)
-Trains multimodal projection adapters and visual cross-attention layers. Operates on image-text pairs formatted in LLaVA or Qwen2-VL visual question answering schemas. Directly bridges into the `termux-vision` mobile inference runtime.
+Trains multimodal projection adapters and visual cross-attention layers. Operates on image-text pairs formatted in LLaVA or Qwen2-VL visual question answering schemas. Directly bridges into the `amv-vision` mobile inference runtime.
 
 ```bash
 amv-train vision-train \
@@ -119,7 +119,7 @@ amv-train vision-train \
 ```
 
 ### 3. Speech-to-Text Whisper LoRA Training (`stt-train`)
-Performs acoustic adaptation on Whisper encoder-decoder attention blocks. Features automated Mel filterbank spectrogram extraction and persistent caching. Natively consumable by `termux-stt`.
+Performs acoustic adaptation on Whisper encoder-decoder attention blocks. Features automated Mel filterbank spectrogram extraction and persistent caching. Natively consumable by `amv-stt`.
 
 ```bash
 amv-train stt-train \
@@ -133,7 +133,7 @@ amv-train stt-train \
 ```
 
 ### 4. Text-to-Speech Style Adaptation LoRA Training (`tts-train`)
-Fine-tunes speaker embedding and acoustic style parameters for neural speech synthesis. Generates low-rank voice profile adapters directly compatible with `termux-tts`.
+Fine-tunes speaker embedding and acoustic style parameters for neural speech synthesis. Generates low-rank voice profile adapters directly compatible with `amv-tts`.
 
 ```bash
 amv-train tts-train \
@@ -165,7 +165,7 @@ amv-train train \
 ```
 
 ### 6. BitNet 1.58-bit Ternary Quantization-Aware Training
-Applies dynamic activation scaling and ternary weight quantization $\{-1, 0, +1\}$ during forward and backward passes. Directly exports quantized kernels compatible with `bitnet.cpp` and `termux-bitnet`.
+Applies dynamic activation scaling and ternary weight quantization $\{-1, 0, +1\}$ during forward and backward passes. Directly exports quantized kernels compatible with `bitnet.cpp` and `amv-bitnet`.
 
 ```bash
 amv-train train \
@@ -242,7 +242,7 @@ To neutralize these constraints:
 > npm install -g @ameva/cluster amv-train
 > ```
 
-When training deep learning models that exceed physical mobile device memory (e.g. 7B+ parameter models, high-resolution diffusion unets, or multimodal projectors), Termux-Train binds heterogeneous smartphone clusters into a unified virtual memory pool. Idle mobile devices contribute unallocated LPDDR RAM via lightweight worker daemons managed by AMEVA Cluster.
+When training deep learning models that exceed physical mobile device memory (e.g. 7B+ parameter models, high-resolution diffusion unets, or multimodal projectors), AMV-Train binds heterogeneous smartphone clusters into a unified virtual memory pool. Idle mobile devices contribute unallocated LPDDR RAM via lightweight worker daemons managed by AMEVA Cluster.
 
 ```
 +-------------------------------------------------------------------------+
@@ -322,7 +322,7 @@ When passing sensitive weights and intermediate activations across untrusted Wi-
 
 ### Python SDK
 ```python
-import termux_train as tt
+import amv_train as tt
 
 # High-level training interface
 trainer = tt.TermuxTrainer(
@@ -363,15 +363,15 @@ console.log(`Diffusion training completed with status: ${result.status}`);
 
 ## Target Runtime Compatibility & Adapter Hub
 
-All adapters generated by Termux-Train are zero-conversion artifacts accompanied by a standardized `_config.json` sidecar.
+All adapters generated by AMV-Train are zero-conversion artifacts accompanied by a standardized `_config.json` sidecar.
 
 | Target Runtime | Adapter Format | Companion Sidecar | Validation Status |
 | :--- | :--- | :--- | :--- |
 | **ComfyUI / Diffusers** | `.safetensors` (UNet CrossAttention) | `adapter_diffusion_config.json` | Validated by automated checks |
 | **llama.cpp / GGUF** | `.safetensors` / GGUF LoRA | `adapter_llm_config.json` | Validated by automated checks |
-| **termux-vision** | `.safetensors` (Multimodal Projector) | `adapter_vision_config.json` | Validated by automated checks |
-| **termux-stt** | `.safetensors` (Whisper Attention) | `adapter_stt_config.json` | Validated by automated checks |
-| **termux-tts** | `.safetensors` (Speaker Style) | `adapter_tts_config.json` | Validated by automated checks |
+| **amv-vision** | `.safetensors` (Multimodal Projector) | `adapter_vision_config.json` | Validated by automated checks |
+| **amv-stt** | `.safetensors` (Whisper Attention) | `adapter_stt_config.json` | Validated by automated checks |
+| **amv-tts** | `.safetensors` (Speaker Style) | `adapter_tts_config.json` | Validated by automated checks |
 | **bitnet.cpp** | `.safetensors` (Ternary 1.58-bit) | `bitnet_config.json` | Validated by automated checks |
 
 ---

package/README.pypi.md

@@ -1,15 +1,15 @@
-# Termux-Train (Python)
+# AMV-Train (Python)
 
-[![PyPI](https://img.shields.io/pypi/v/termux-train.svg?style=flat-square&color=0369a1)](https://pypi.org/project/termux-train/)
-[![Python](https://img.shields.io/pypi/pyversions/termux-train.svg?style=flat-square)](https://pypi.org/project/termux-train/)
-[![License](https://img.shields.io/badge/License-Apache_2.0-004499.svg?style=flat-square)](https://github.com/uno-km/termux-train)
+[![PyPI](https://img.shields.io/pypi/v/amv-train.svg?style=flat-square&color=0369a1)](https://pypi.org/project/amv-train/)
+[![Python](https://img.shields.io/pypi/pyversions/amv-train.svg?style=flat-square)](https://pypi.org/project/amv-train/)
+[![License](https://img.shields.io/badge/License-Apache_2.0-004499.svg?style=flat-square)](https://github.com/uno-km/amv-train)
 
 > Unified Multimodal On-Device Deep Learning & LoRA Training Framework for Android Termux with 6-Modality Adapters (LLM, Diffusion, VLM, STT, TTS, BitNet), GPU Slicing, and 44GB Disaggregated Cluster Virtual RAM Pooling.
 
 ## Installation
 
 ```bash
-pip install termux-train
+pip install amv-train
 
 # For multi-device distributed 44GB Virtual RAM Pooling (Required):
 pip install ameva-cluster
@@ -18,7 +18,7 @@ pip install ameva-cluster
 ## Python Quickstart
 
 ```python
-import termux_train as tt
+import amv_train as tt
 
 # 1. Diffusion LoRA Training
 tt.diffusion.train_diffusion_lora(
@@ -48,10 +48,10 @@ To train models exceeding single-device physical memory across multiple Android
 ameva-cluster worker --port 50052 --guard-band 300
 
 # Step 2: Probe fleet from coordinator phone (e.g. S25)
-termux-train cluster-probe --fleet 192.168.1.101:50052,192.168.1.102:50052
+amv-train cluster-probe --fleet 192.168.1.101:50052,192.168.1.102:50052
 
 # Step 3: Run distributed pipeline training on aggregated 44GB virtual RAM pool
-termux-train train --model tiny-transformer --data ./corpus.txt \
+amv-train train --model tiny-transformer --data ./corpus.txt \
   --virtual-ram-pool 192.168.1.101:50052,192.168.1.102:50052 \
   --vocab-slice 4096 --chunk-layers 2 --backend vulkan
 ```
@@ -60,24 +60,24 @@ termux-train train --model tiny-transformer --data ./corpus.txt \
 
 ```bash
 # Diffusion LoRA
-termux-train diffusion-train --image-dir ./images --output ./adapter.safetensors --epochs 5
+amv-train diffusion-train --image-dir ./images --output ./adapter.safetensors --epochs 5
 
 # Vision VLM LoRA
-termux-train vision-train --data ./vlm.jsonl --output ./vlm_adapter.safetensors --epochs 3
+amv-train vision-train --data ./vlm.jsonl --output ./vlm_adapter.safetensors --epochs 3
 
 # Whisper STT LoRA
-termux-train stt-train --data ./audio.jsonl --output ./stt_adapter.safetensors --epochs 4
+amv-train stt-train --data ./audio.jsonl --output ./stt_adapter.safetensors --epochs 4
 
 # TTS Style LoRA
-termux-train tts-train --data ./speech.jsonl --output ./tts_adapter.safetensors --epochs 5
+amv-train tts-train --data ./speech.jsonl --output ./tts_adapter.safetensors --epochs 5
 
 # LLM LoRA with GPU Slicing
-termux-train train --model tiny-transformer --data ./corpus.txt --vocab-slice 4096 --chunk-layers 2
+amv-train train --model tiny-transformer --data ./corpus.txt --vocab-slice 4096 --chunk-layers 2
 ```
 
 ## Documentation & Repository
 - Official Documentation: https://uno-km.vercel.app/lib/train/
-- GitHub Repository: https://github.com/uno-km/termux-train
+- GitHub Repository: https://github.com/uno-km/amv-train
 
 ## License
 Apache-2.0 License. Copyright (c) 2026 Eunho Kim (@uno-km).

package/bin/cli.js

@@ -5,7 +5,7 @@
  */
 
 /**
- * termux-train: Node.js Global CLI Binary.
+ * amv-train: Node.js Global CLI Binary.
  * High-performance on-device Deep Learning & LoRA Training Framework.
  * Open-Source under Apache License 2.0.
  */
@@ -51,8 +51,8 @@ function hasFlag(flag, alias = null) {
 }
 
 function printHelp() {
-  console.log(`termux-train CLI v${version} (Unified Node.js & Python Dual Engine)`);
-  console.log('Usage: termux-train <command> [options]\n');
+  console.log(`amv-train CLI v${version} (Unified Node.js & Python Dual Engine)`);
+  console.log('Usage: amv-train <command> [options]\n');
   console.log('Core Commands:');
   console.log('  doctor                           Inspect device hardware, RAM tier, and Vulkan GPU');
   console.log('  check                            Run self-diagnostic mathematical checks across backends');
@@ -87,7 +87,7 @@ function printHelp() {
 
 async function main() {
   if (hasFlag('-v') || hasFlag('--version')) {
-    console.log(`termux-train ${version}`);
+    console.log(`amv-train ${version}`);
     process.exit(0);
   }
 
@@ -103,10 +103,10 @@ async function main() {
     if (isJson) {
       console.log(JSON.stringify(rep, null, 2));
     } else {
-      console.log('=== termux-train Diagnostic Doctor (Node.js Engine) ===');
+      console.log('=== amv-train Diagnostic Doctor (Node.js Engine) ===');
       console.log(`  Platform       : ${rep.platform.system} (${rep.platform.arch}) | Android/Termux: ${rep.platform.isAndroid || rep.platform.isTermux}`);
       console.log(`  Hardware Tier  : ${rep.hardware.tier} (RAM: ${rep.hardware.totalRamMb}MB | Cores: ${rep.hardware.cpuCores})`);
-      console.log(`  Python Runtime : ${rep.python.available ? rep.python.version : 'NOT FOUND'} (${rep.python.moduleInstalled ? 'termux_train installed' : 'module missing'})`);
+      console.log(`  Python Runtime : ${rep.python.available ? rep.python.version : 'NOT FOUND'} (${rep.python.moduleInstalled ? 'amv_train installed' : 'module missing'})`);
       console.log(`  Vulkan GPU     : ${rep.vulkan.status}`);
       console.log(`  Preset Config  : Max LoRA Rank r=${rep.recommendedTrainingConfig.loraRank} | Batch Size=${rep.recommendedTrainingConfig.batchSize}`);
       console.log(`  Overall Status : ${rep.status}`);
@@ -122,7 +122,7 @@ async function main() {
       if (isJson) {
         console.log(JSON.stringify(res, null, 2));
       } else {
-        console.log(`=== termux-train Benchmark (Dimension: ${res.dimension}) ===`);
+        console.log(`=== amv-train Benchmark (Dimension: ${res.dimension}) ===`);
         console.log(`  Backend                   : ${res.backend}`);
         console.log(`  Forward GEMM Latency      : ${res.gemmLatencyMs} ms`);
         console.log(`  Full Autograd Step (F+B)  : ${res.autogradStepLatencyMs} ms`);
@@ -208,7 +208,7 @@ async function main() {
       console.error('[ERROR] Python runtime is required to execute native training pipelines.');
       process.exit(1);
     }
-    const pyArgs = ['-m', 'termux_train.cli', ...args];
+    const pyArgs = ['-m', 'amv_train.cli', ...args];
     const res = spawnSync(pyCmd, pyArgs, { stdio: 'inherit' });
     process.exit(res.status || 0);
   }

package/index.d.ts

@@ -3,7 +3,7 @@
  * SPDX-License-Identifier: Apache-2.0 OR Commercial-License
  */
 /**
- * Type declarations for termux-train Node.js SDK
+ * Type declarations for amv-train Node.js SDK
  * Open-Source under Apache License 2.0.
  */
 
@@ -46,7 +46,7 @@ export interface RecommendedTrainingConfig {
 }
 
 export interface DoctorReport {
-  framework: 'termux-train';
+  framework: 'amv-train';
   version: string;
   platform: DoctorPlatformInfo;
   hardware: DoctorHardwareInfo;

package/index.js

@@ -3,7 +3,7 @@
  * SPDX-License-Identifier: Apache-2.0 OR Commercial-License
  */
 /**
- * termux-train: Native On-Device Deep Learning & LoRA Training Framework
+ * amv-train: Native On-Device Deep Learning & LoRA Training Framework
  * Dual-Engine (Python & Node.js/TypeScript) Native Module for Android Termux & ARM64.
  * Open-Source under Apache License 2.0.
  */

package/lib/benchmark.js

@@ -3,9 +3,9 @@
  * SPDX-License-Identifier: Apache-2.0 OR Commercial-License
  */
 /**
- * termux-train: Benchmark Suite
+ * amv-train: Benchmark Suite
  * High-performance GEMM & DAG Autograd Latency & Throughput Benchmark.
- * All benchmark execution uses stdin JSON IPC via termux_train.runtime.benchmarker
+ * All benchmark execution uses stdin JSON IPC via amv_train.runtime.benchmarker
  * — zero python -c inline script injection.
  * Open-Source under Apache License 2.0.
  */
@@ -45,7 +45,7 @@ async function runBenchmarkAsync(options = {}) {
   const cfg = JSON.stringify({ dim, iters });
 
   return new Promise((resolve, reject) => {
-    const proc = spawn(pyCmd, ['-m', 'termux_train.runtime.benchmarker', '--stdin-json'], {
+    const proc = spawn(pyCmd, ['-m', 'amv_train.runtime.benchmarker', '--stdin-json'], {
       stdio: ['pipe', 'pipe', 'pipe']
     });
 
@@ -60,7 +60,7 @@ async function runBenchmarkAsync(options = {}) {
 
     proc.on('close', (code) => {
       if (code !== 0) {
-        if (stderrAcc.includes('No module named termux_train')) {
+        if (stderrAcc.includes('No module named amv_train')) {
           return reject(new TermuxTrainModuleNotFoundError());
         }
         return reject(new Error(`Benchmark process failed (exit code ${code}): ${stderrAcc}`));
@@ -90,7 +90,7 @@ function runBenchmark(options = {}) {
 
   const cfg = JSON.stringify({ dim, iters });
 
-  const res = spawnSync(pyCmd, ['-m', 'termux_train.runtime.benchmarker', '--stdin-json'], {
+  const res = spawnSync(pyCmd, ['-m', 'amv_train.runtime.benchmarker', '--stdin-json'], {
     input: cfg,
     stdio: ['pipe', 'pipe', 'pipe'],
     encoding: 'utf-8'
@@ -98,7 +98,7 @@ function runBenchmark(options = {}) {
 
   if (res.status !== 0) {
     const err = res.stderr || res.stdout || '';
-    if (err.includes('No module named termux_train')) {
+    if (err.includes('No module named amv_train')) {
       throw new TermuxTrainModuleNotFoundError();
     }
     throw new Error(`Benchmark failed: ${err}`);

package/lib/doctor.js

@@ -3,7 +3,7 @@
  * SPDX-License-Identifier: Apache-2.0 OR Commercial-License
  */
 /**
- * termux-train: Hardware & Training Capacity Diagnostic Doctor
+ * amv-train: Hardware & Training Capacity Diagnostic Doctor
  * Production-grade heuristics with free RAM safety margin calculation.
  * Open-Source under Apache License 2.0.
  */
@@ -96,7 +96,7 @@ function runDoctor(options = {}) {
 
       const checkMod = spawnSync(
         pyCmd,
-        ['-c', 'import termux_train as tt; b = tt.get_backend("vulkan") if "vulkan" in tt.available_backends() else None; print(getattr(b, "is_vulkan_active", False))'],
+        ['-c', 'import amv_train as tt; b = tt.get_backend("vulkan") if "vulkan" in tt.available_backends() else None; print(getattr(b, "is_vulkan_active", False))'],
         { stdio: 'pipe', encoding: 'utf-8' }
       );
       if (!checkMod.error && checkMod.status === 0) {
@@ -107,7 +107,7 @@ function runDoctor(options = {}) {
   }
 
   return {
-    framework: 'termux-train',
+    framework: 'amv-train',
     version: require('../package.json').version,
     platform: {
       system: process.platform,

package/lib/errors.js

@@ -3,7 +3,7 @@
  * SPDX-License-Identifier: Apache-2.0 OR Commercial-License
  */
 /**
- * termux-train: Standard Error Definitions
+ * amv-train: Standard Error Definitions
  * Open-Source under Apache License 2.0.
  */
 
@@ -24,7 +24,7 @@ class PythonRuntimeNotFoundError extends TermuxTrainError {
 }
 
 class TermuxTrainModuleNotFoundError extends TermuxTrainError {
-  constructor(msg = 'termux_train Python module is not installed or importable. Run "pip install termux-train".') {
+  constructor(msg = 'amv_train Python module is not installed or importable. Run "pip install amv-train".') {
     super(msg);
   }
 }

package/lib/multimodal.js

@@ -3,7 +3,7 @@
  * SPDX-License-Identifier: Apache-2.0 OR Commercial-License
  */
 /**
- * termux-train: Multimodal & Distributed Training Node.js Bridge
+ * amv-train: Multimodal & Distributed Training Node.js Bridge
  * Provides asynchronous JavaScript SDK bindings for Diffusion, Vision VLM,
  * Whisper STT, TTS Speaker Adaptation, and Cluster Orchestration.
  * Open-Source under Apache License 2.0.
@@ -22,7 +22,7 @@ function _executePythonCli(subcommand, optArgs = {}) {
       return reject(new PythonRuntimeNotFoundError('Python runtime required for multimodal training execution.'));
     }
 
-    const cliArgs = ['-m', 'termux_train.cli', subcommand];
+    const cliArgs = ['-m', 'amv_train.cli', subcommand];
     for (const [k, v] of Object.entries(optArgs)) {
       if (v === null || v === undefined) continue;
       const flagName = k.replace(/([A-Z])/g, '-$1').toLowerCase();

package/lib/trainer.js

@@ -3,7 +3,7 @@
  * SPDX-License-Identifier: Apache-2.0 OR Commercial-License
  */
 /**
- * termux-train: Production-Grade On-Device Training Orchestrator
+ * amv-train: Production-Grade On-Device Training Orchestrator
  * True Dual-Engine LoRA, Transformer & MLP Training Pipeline with SafeTensors I/O.
  * Open-Source under Apache License 2.0.
  */
@@ -192,7 +192,7 @@ class TermuxTrainer extends EventEmitter {
       this._lineBuffer = '';
       this._stoppedEmitted = false;
 
-      this.process = spawn(pyCmd, ['-m', 'termux_train.runtime.runner', '--stdin-json'], {
+      this.process = spawn(pyCmd, ['-m', 'amv_train.runtime.runner', '--stdin-json'], {
         stdio: ['pipe', 'pipe', 'pipe']
       });
 

package/package.json

@@ -1,11 +1,11 @@
 {
   "name": "amv-train",
-  "version": "2.1.0",
+  "version": "3.0.0",
   "description": "On-device deep learning & LoRA training framework utilizing device resources for Android Termux & ARM64",
   "main": "index.js",
   "types": "index.d.ts",
   "bin": {
-    "termux-train": "bin/cli.js",
+    "amv-train": "bin/cli.js",
     "tt": "bin/cli.js"
   },
   "files": [
@@ -42,10 +42,10 @@
   "homepage": "https://uno-km.vercel.app/lib/train/",
   "repository": {
     "type": "git",
-    "url": "git+https://github.com/uno-km/termux-train.git"
+    "url": "git+https://github.com/uno-km/amv-train.git"
   },
   "bugs": {
-    "url": "https://github.com/uno-km/termux-train/issues"
+    "url": "https://github.com/uno-km/amv-train/issues"
   },
   "dependencies": {},
   "optionalDependencies": {