Skip to content
View Qiyuanqiii's full-sized avatar
:dependabot:
GPT 6 Astra Ultra Working
:dependabot:
GPT 6 Astra Ultra Working

Organizations

@CWNU-Open-Source-Community

Block or report Qiyuanqiii

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Qiyuanqiii/README.md

Animated profile banner

Typing introduction

祈愿 Qiii

Cybersecurity student based in Sichuan, China.
I build local-first agent systems, reliability tooling, and open-source fixes that are auditable, testable, and maintainable.

📊 At a Glance

Profile views Followers 145 public pull requests, 105 merged

📬 Connect

X @Qiyuanqiii Bilibili Qiyuanqiii LeetCode qiyuanqiii Email 2297740147@qq.com

🤖 LLM Toolchain

GPT-6 Astra · Claude Opus 5 · Grok 4.6 · Kimi K3


How I work with AI — my approach to planning, agent collaboration, independent review, and human responsibility. Full text follows.

Read as text

How I work with AI

I bring AI into every stage of a project, from defining the problem to verifying the result. My role is to give that collaboration direction, structure, and a clear standard of completion.

I start with a goal and its boundaries. Together with the most capable AI available for the task, I research the problem, explore existing products and reusable work, and capture the plan in a GitHub issue. We then break it into smaller issues, each with explicit dependencies, constraints, and acceptance criteria.

Development becomes a collaboration between agents. Several handle implementation, tests, and supporting evidence. My strongest reviewer coordinates their work and critically assesses whether the code is robust, maintainable, and ready to ship. Another agent independently examines those judgments—questioning assumptions, framework choices, and unnecessary complexity. Both the implementation and the reasoning behind its approval remain open to revision.

I stay close to the work throughout: following progress, clarifying intent, keeping scope under control, and resolving consequential trade-offs. I remain responsible for the direction and the outcome.

Software takes many steps. Each conversation can clarify a requirement, challenge an assumption, or produce a small, verified improvement. I believe lasting work grows through the patient accumulation of those steps.

“A tree that fills the arms grows from a tiny shoot; a terrace nine storeys high rises from a mound of earth.”

Tao Te Ching, Chapter 64

Per aspera ad astra.

Tech Stack

Animated technology stack: Go, Rust, Python, TypeScript, React, Node.js, Wails, Codex, and Claude Code

Languages · Go · Rust · Python · TypeScript  │  App Development · React · Node.js · Wails  │  AI Coding · Codex · Claude Code

Find me in open source

You can find me in the following open-source community.

OpenCodeReview — human contributor rank #6 / 162. Issue assessment, PR review, personal-branch SVN support, and validation and fixes. Full text follows.

Read my contributions as text

Human contributor rank: #6 / 162

My contributions:

  • Issue assessment. I assess the rationale, feasibility, and scope of proposed changes.
  • PR review. I review pull requests for security vulnerabilities, robustness, and failure handling.
  • SVN support. I develop Subversion (SVN) support on my personal branch, separate from upstream.
  • Validation & fixes. I validate and fix data-handling vulnerabilities, boundary-condition errors, and overflow issues.

Snapshot: 8 September 2026. Rank excludes bot accounts; 162 is GitHub's displayed contributor total, not a human-only count. Contributors · My pull requests

GitHub Activity

GitHub achievement trophies

GitHub stats Most used languages

GitHub contribution streak

Contribution activity graph

Issue and pull request lifecycle statistics

Isometric contribution calendar

Contribution snake animation

🔧 About this profile

The streak, trophy, contribution calendar, and snake animations on this page are regenerated automatically by GitHub Actions and committed back to this repository. Only the README and the assets are hand-maintained.

Build the system. Verify the boundary. Ship the fix.

Footer

Pinned Loading

  1. CWNU-Open-Source-Community/Traverse-Board CWNU-Open-Source-Community/Traverse-Board Public

    (未完成)这是一个让执行可以恢复、让结果可以追溯的 Agent Harness。在深海之上,网页不可信,模型输出不可信,工具输出不可信,过去的记忆即便真实,也未必适用于现在。不要相信“它说自己做过什么”,去回顾它留下的航迹。

    Go 179 4

  2. Jump-Terminator Jump-Terminator Public

    Jump Terminator Android 跨应用跳转控制工具(S0 技术可行性验证)

    Python 2

  3. TencentCloud/TencentDB-Agent-Memory TencentCloud/TencentDB-Agent-Memory Public

    TencentDB Agent Memory is a team-level memory hub for AI Agents — turning conversations, docs, and code into four reusable memory assets (Chat Memory, Skill, LLM-Wiki, Code-Graph) that are governed…

    TypeScript 26.2k 2.5k

  4. Tencent/VulnGym Tencent/VulnGym Public

    VulnGym: A Real-World, Project-Level Vulnerability Benchmark for White-Box Vulnerability-Hunting Agents

    245 45

  5. CWNU-Open-Source-Community/SCSWiki CWNU-Open-Source-Community/SCSWiki Public

    SCSWiki 是一个学生维护的非官方西华师范大学计算机学院知识库,目标是把可复用的学习经验、工具入门、内容模板和贡献规范整理到一个可搜索、可审阅、可持续维护的地方,让新生、在读同学更快找到可靠起点。

    TypeScript 8 3

  6. alibaba/open-code-review alibaba/open-code-review Public

    Fast, efficient, battle-tested at Alibaba's scale. Hybrid architecture code review tool: deterministic pipelines + LLM Agent, precise line-level comments, built-in multi-language ruleset (NPE, thre…

    Go 22.1k 1.6k