Skip to content

About

AloneAIOR is an automated game control system for the popular game TalesRunner. This project aims to provide a fully automated experience by leveraging various Windows API calls and custom routines to interact with the game window, simulate user inputs, capture screen data, and automate in-game actions.

Resources

Security policy

Stars

5 stars

Watchers

1 watching

Forks

Repository files navigation

AloneAIO Framework

Vision-driven Tales Runner automation for Windows

Automated workflows, route execution, OCR, adaptive recovery, and multi-backend input in one desktop runtime.

Download ZIP · Getting Started · Features · Architecture · 繁體中文 · Discord

Windows x64 · .NET Framework 4.8 · Tales Runner · Windows Forms


Overview

AloneAIO is a Windows desktop automation runtime for Tales Runner. It watches the game through computer vision, identifies the current state, runs the selected workflow, and applies recovery logic when progress stalls or the game leaves the expected state.

The runtime combines everyday automation features with an engineering-focused execution model: routes are stateful, input is centrally managed, visual checks are reusable, and every long-running workflow has explicit stop and cleanup paths.

Highlights

Capability Description
Automation Full workflow coordination Coordinates game launch, login, rooms, maps, events, fishing, and repeated AFK operation.
Vision State detection and OCR Uses pixel evidence, template matching, and OCR to interpret the live game screen.
Routes Phase-based map execution Runs structured map routes with states, triggers, combos, and recovery zones.
Recovery Adaptive correction Detects stalled progress, retries safely, and escalates to controlled recovery.
Input Configurable input modes Resolves logical game actions through the selected input backend.
Languages Multi-language UI support Includes English, Traditional Chinese, Simplified Chinese, Korean, and Thai resources.

Features

Automation Modes

  • Automatic event workflows.
  • Automatic map rotation and custom map routes.
  • Fishing automation.
  • Guild and room-oriented workflows.
  • Daily task and item-use coordination.
  • Quiz and mini-game assistance using OCR and text matching.
  • Multi-account workflow support.

Vision Runtime

  • Game-state classification from one captured frame.
  • Region-based pixel and color matching.
  • Image template matching.
  • OCR for numbers, text, and supported language content.
  • Batched visual checks for repeated state detection.
  • Capture-health monitoring and failure recovery.

Route Runtime

  • Versioned route documents.
  • State, trigger, macro, and recovery phases.
  • Configurable directional actions and combos.
  • Pause, resume, stop, and terminal input cleanup.
  • Progress-aware stage transitions.
  • Runtime-only route logging.

Adaptive Recovery

AloneAIO treats movement as a feedback loop rather than a fixed macro. It compares recent progress with expected movement, detects repeated stalls, selects a correction, verifies the result, and stops safely when recovery limits are reached.

Getting Started

Requirements

  • Windows x64.
  • .NET Framework 4.8.
  • Tales Runner installed locally.
  • 100% Windows display scaling for the expected capture coordinates.
  • Administrator rights when required by the selected input mode.

Install

  1. Download the repository ZIP.
  2. Extract the complete archive.
  3. Open AloneAIOR/bin/Debug/.
  4. Edit Alone.ini for your local Tales Runner path and preferred mode.
  5. Run AloneAIOR.exe.

Keep the complete bin/Debug directory together. The executable uses the native libraries and OCR data stored under lib/.

Default Configuration

AloneAIOR/bin/Debug/Alone.ini is included as the default configuration.

Common settings:

Section Setting Purpose
Setting TRFile Local path to the Tales Runner executable.
Setting FullAuto Starts the configured automation workflow automatically.
AFKmode AFKmode Selects the main unattended mode.
AFKmode Map / AutoMap Selects fixed or automatic map behavior.
Input Mode Selects the configured input mode.
Delay SlowPC Enables more conservative timing for slower systems.
RoomSetting Channel / CreateRoom Controls room and channel behavior.

Account and Password are blank in the tracked default file. Keep personal credentials local and do not commit them.

How It Works

flowchart LR
    CAPTURE[Capture Game State] --> DETECT[Classify Visual Evidence]
    DETECT --> DECIDE[Select Workflow Action]
    DECIDE --> INPUT[Dispatch Logical Input]
    INPUT --> VERIFY[Verify Progress]
    VERIFY -->|progress| CAPTURE
    VERIFY -->|stalled| RECOVER[Apply Recovery Policy]
    RECOVER --> CAPTURE
Loading

At runtime, the application repeatedly observes the game, decides what the active workflow requires, sends the corresponding logical actions, and verifies the result. Vision, input, route execution, and recovery remain separate responsibilities so each can be maintained independently.

Runtime Layout

AloneAIOR/bin/Debug/
|-- AloneAIOR.exe
|-- Alone.ini
`-- lib/
    |-- x64/
    |   |-- leptonica-1.80.0.dll
    |   |-- tesseract41.dll
    |   `-- winhid.dll
    `-- tessdata/
        |-- chi_tra.traineddata
        `-- eng.traineddata

Architecture For Developers

The repository includes a readable architecture reference beside the runtime files. It shows the major responsibility areas, runtime flows, module ownership, and a small set of interface contracts.

The architecture reference is intended for reading and design study; the downloadable runtime remains under AloneAIOR/bin/Debug.

Roadmap

The current research direction explores adaptive route selection based on progress evidence and bounded action scoring. See docs/AI-Pathfinding.md.

Support

About

AloneAIOR is an automated game control system for the popular game TalesRunner. This project aims to provide a fully automated experience by leveraging various Windows API calls and custom routines to interact with the game window, simulate user inputs, capture screen data, and automate in-game actions.

Resources

Security policy

Stars

5 stars

Watchers

1 watching

Forks

Contributors

Languages