01. Who we are and what this policy covers
iPlugin Junqi (“the plugin”) is a public, face-up two-player Luzhanqi (军棋) game for ChatGPT. It is published and operated by Trustflow as part of iPlugin. In this policy, “we,” “us,” and “our” refer to that publisher and operator.
This policy applies to the plugin listing, the fullscreen 12 × 5 field, the MCP server at https://iplugin.dev/junqi/mcp, the review video, and the Junqi support and data-deletion pages. It does not apply to ChatGPT itself, OpenAI’s handling of your conversations, or any third-party website. OpenAI’s privacy policy and terms govern the ChatGPT service.
The plugin has no account system. You do not create an iPlugin account, sign in with Google, authorize OAuth access, pay for the plugin, or provide an API key to play. The public field and the game tools are available without a login.
02. At a glance
The table below summarizes the categories of personal data the plugin processes, the purposes of use, the categories of recipients, and retention timelines. The sections that follow explain each row in more detail.
| Information | Why we process it | Recipients | Retention |
|---|---|---|---|
| Game state: game identifier, 12 × 5 board, fifty pieces and their ranks, moves, road and railway paths, battles, captured pieces, turn, first player, difficulty, status, winner, resignation flag, army counts, and setup seed | Run the match, validate every road step and railway run, resolve battles, enforce camp, headquarters and flag rules, and return the current position | ChatGPT (as tool results); our hosting infrastructure | Server memory only; deleted when the game is replaced or the service restarts. Not written to a database. |
| Model-turn prompt: game identifier, board snapshot, both army summaries, difficulty, and legal move list | Ask the current ChatGPT model to choose a legal blue move in the existing conversation | OpenAI, as part of your ChatGPT conversation | Governed by OpenAI’s conversation retention settings. We do not keep a separate copy. |
| Local preferences: difficulty, AI First, and whether legal-move hints are shown | Remember your board settings between visits | Stored only in your browser; not sent to us | Until you clear site data for the plugin |
| Technical request data: IP address, request time, path, response status, and error details | Security, rate limiting, abuse prevention, and reliability | Our hosting and network infrastructure | A short operational period, then rotated or deleted; longer only if needed for security or legal reasons |
03. Information we process
The categories of personal data we process are gameplay state, a model-turn prompt, local browser preferences, and technical request data. Each category is described below.
3.1 Game state
When you start or continue a game, the plugin processes a game identifier, the 12 × 5 board, the positions and types of the 25 red and 25 blue pieces, each move’s from and to squares, the road or railway path, the number of squares travelled, any battle and its result, captured pieces, whose turn it is, who moved first, the selected difficulty, the game status, the winner and win reason, whether a player resigned, the remaining army counts, and the setup seed used for the deployment.
The game is face-up: both armies are visible. There is no hidden deployment, no hidden piece and no referee. The plugin uses the game state to validate legal moves, apply the camp and headquarters rules, resolve battles, detect a flag capture or a side with no legal move, and return the current position. Games are held in the server process memory. They are not written to the iPlugin PostgreSQL database, and a game may disappear when the service is redeployed, restarted, or replaces the active game. We do not keep a history of past games and do not create a player profile.
3.2 The model’s turn
After a red move, or when the game begins with AI First, the board asks the current ChatGPT model for the next blue move. That request contains the game identifier, a board snapshot, both army summaries, the difficulty, and the list of legal moves with their sources, destinations, roads or railways, and capture outcomes. ChatGPT returns one legal move, which the plugin applies through its game tools.
This prompt is part of your ChatGPT conversation. OpenAI processes it under its policies and your account settings. We do not receive your full conversation, unrelated prompts, files, contacts, calendar, browsing history, or ChatGPT credentials. We only receive the arguments passed to the plugin’s tools and the game state the plugin itself creates.
3.3 Browser preferences
The board stores two settings in your browser’s localStorage:
junqi:settings— the selected difficulty and whether AI First is enabled;junqi:show-hints— whether legal destinations are shown.
These values stay on your device until you clear site data for the plugin. They are not sent to us and are not used to build a profile. The active game itself is not stored in localStorage.
3.4 Technical and security data
Our web server and hosting infrastructure may process ordinary request data such as IP address, request time, requested path, response status, client version, and error details for security, rate limiting, abuse prevention, and reliability. The plugin application does not write this information to its own database. Because the plugin has no accounts, these records are not linked to an iPlugin profile.
04. How we use information
We use the information described above only for the following purposes:
- Provide the game: open the field, create a match, deal a legal deployment, validate road steps and railway runs, enforce turn order, resolve battles, apply camp and headquarters rules, detect a flag capture or no-move finish, apply undo and resignation, and return the current board to ChatGPT.
- Coordinate the model’s turn: give the current ChatGPT model the current board, legal moves, army summaries, and difficulty so it can choose a legal blue move in the existing conversation.
- Remember your settings: keep difficulty, AI First, and hints preferences on your device.
- Protect the service: rate-limit requests, investigate errors, prevent abuse, and maintain availability.
- Comply with the law: respond to lawful requests, enforce our terms, and protect the rights and safety of users and the service.
We do not use game data for advertising, behavioral profiling, or sale, and we do not use it to train models. The plugin does not call a separate model API; the current ChatGPT model chooses every blue move through the game tools. There is no Junqi engine, opening book, learning system, hidden piece information, or hidden move selector.
05. What we do not collect
iPlugin Junqi does not ask for, and does not intentionally collect, any of the following:
- your name, email address, phone number, or postal address;
- your Google account, contacts, calendar, Drive files, or other Google data;
- precise location, payment card details, government identifiers, health information, or other sensitive personal data;
- passwords, API keys, OAuth tokens, or other authentication secrets;
- advertising identifiers, cross-site tracking data, or marketing profiles;
- your private chats, files, or content unrelated to the Junqi match.
Please do not send any of these to us through ChatGPT, email, or support requests. If you include sensitive information in a ChatGPT conversation, OpenAI’s policies and your account settings govern that content.
07. Retention
- Game state: kept in server memory until the game is replaced or the service restarts or is redeployed. It is not written to a database and is not included in database backups.
- Technical logs: kept for a short operational period, then rotated or deleted. We may retain specific records longer if needed to investigate abuse, resolve a security incident, or comply with law.
- Browser preferences: kept in your browser until you clear site data for the plugin.
- ChatGPT conversation: retained by OpenAI according to your ChatGPT account settings and OpenAI’s policies. We cannot delete OpenAI conversation records.
- Email correspondence: if you contact us, we keep your message and our reply only as long as needed to respond and for reasonable recordkeeping, then delete it.
Because the plugin has no account or durable server profile, there is no account-deletion step. See Your choices and rights and our data deletion page for the controls that are available.
08. Security
We use HTTPS/TLS for the public service, validate every move, railway path, camp entry, headquarters entry, battle, undo and resignation on the server, restrict the board to 60 valid squares, and keep no account credentials or long-lived secrets. Access to the service infrastructure is limited, dependencies are kept current, and we review security issues when they are reported.
No method of transmission or storage is completely secure. Please avoid sharing sensitive personal information in prompts or support messages. If you believe you have found a vulnerability, contact us at micromvp@gmail.com with enough detail to reproduce the issue; please do not include other users’ data.
09. Your choices and rights
You can start a new game to replace the active game state, use undo or resign to change or end the current match, clear site data for the plugin to remove the local preferences, and use OpenAI’s controls to manage or delete ChatGPT conversation content. These are the controls offered to you. Because there is no iPlugin account, there is no profile to review or delete on our side.
Depending on where you live, you may have rights to access, correct, delete, restrict, or object to the processing of your personal data, to receive a portable copy, to withdraw consent, and to lodge a complaint with your local data-protection authority. You also have the right not to be discriminated against for exercising these rights.
To make a request, email micromvp@gmail.com or use the support page. Because we do not maintain accounts, we may not be able to identify you in our records; we will explain what we can and cannot do. We may ask for limited information to locate a specific log entry, and we will respond within the time required by applicable law.
California notice: in the past 12 months, we have collected the categories described in section 03 (gameplay state, the model-turn prompt, local preferences, and technical request data). We do not sell personal information, do not share it for cross-context behavioral advertising, and do not collect sensitive personal information. California residents may request to know, delete, or correct personal information, may use an authorized agent, and will not be discriminated against for exercising these rights.
10. Children
iPlugin Junqi is intended for a general audience aged 13 and older. It is not directed to children under 13, and we do not knowingly collect personal information from children under 13. If you believe a child under 13 has used the plugin, contact us and we will delete any information we control. ChatGPT conversation content is handled through OpenAI’s account and privacy controls.
11. International transfers
We and our service providers may process information in the United States and other countries. Privacy laws in those countries may differ from the laws where you live. Where required, we rely on appropriate safeguards such as the European Commission’s Standard Contractual Clauses, the UK Addendum, or equivalent safeguards recognized under Swiss law. You can contact us for more information about the safeguards that apply.
12. Changes to this policy
We will update this page and its effective date when our practices change. Material changes will be communicated through the plugin listing or the product page. If a change requires your consent under applicable law, we will ask for it before the change applies. Continued use of the plugin after the effective date means you accept the updated policy.
13. Contact
iPlugin Junqi is published and operated by Trustflow (iPlugin). For privacy questions, data requests, or security reports, email micromvp@gmail.com or use the Junqi support page.
Please do not send passwords, access tokens, or private ChatGPT conversation content. If your request concerns a specific game, include the approximate time of the request rather than the full conversation.