🏆 Notable contributions
Popular open-source repos they've shipped to (commits + PRs)
Score breakdown
🛠 Featured work
Their own popular and pinned repositories
🚀 多Agent协作智能知识库系统 — 五层可编排Agent架构 + RAG全链路 + 统一ToolRegistry，Spring Boot + FastAPI + React，核心亮点是 五层可编排 Agent 架构——Interface、Orchestrator、Tool、Memory、Evaluation 五层解耦，支持意图识别→问题改写→检索→充分性判断→答案生成的完整工作流。系统内置 Router / Retrieval / Ops / Inspection 多 Agent 协作，通过统一 Tool Registry 管理 AI 能力，支持 runId/traceId 全链路追踪、
Python一站式查询全国高校历年高考、研究生招生分数线、录取人数及政策文件，支持离线使用，数据打包在 SQLite 中，无需联网。
PythonShopAgent-X 是一个面向电商场景的多模态智能导购 AI Agent 系统。用户可以通过自然语言对话、拍照识图、语音输入等方式与 AI 交互，获取个性化商品推荐、购物车管理等完整购物体验。
Python知学汇项目是一款大型在线学习网站，基于 **SpringCloud Alibaba** 微服务技术栈，集成在线教育、社交、电商等服务。引入 **Spring AI 框架**构建智能 **Agent** 服务，为学生端提供个性化课程推荐，为管理端提供自动化运营助手（周报/月报生成、数据查询）。学生端核心业务包括视频点播、积分排行、互动问答、课程购买等；管理端核心业务包括用户管理、课程管理、优惠券管理等。
Java📚 ExamPointAI —— 大学生自己的驾考宝典 AI驱动的智能刷题平台，将老师发放的Word/PDF题库在30秒内转化为可刷题、可统计、可复习的智能刷题系统
Python日/月计划与成果验收系统 | 基于SpringBoot3 + 状态机(FSM) + 策略模式的审批闭环，含RBAC数据权限隔离。
Java🧬 Stack & domains
🧬 Most similar developers
Closest profile, nearby score
🔥 Full roast
🔥 41 stars across 10 original repos, 83% of external PRs are test refactors for Dify, and 39 verified core PRs can't hide that your top home project is a 5-star shop agent with zero traction. Your GitHu
EvanYao826 — 75.70/100 · SOLID (Solid · Trustworthy)
TL;DR: Reliable external contributor with a solid track record across popular open source projects, but his own project portfolio is starved for traction and his PR profile is dominated by low-impact test refactors for a single big repo.
| Dimension | Score | Notes |
|---|---|---|
| Account maturity | 5.9/10 | Registered 2.5 years ago, active for 3 contribution years, last activity 0 days ago — young account, but consistent recent activity keeps it from being a ghost profile. |
| Original project quality | 10.7/18 | 10 original repos, 41 total stars across all of them, top repo (AgentCraft) scored 0.9 for quality but only 27 stars, best quality repo (china-university-admission) has 8 stars — your own projects barely move the needle, even the decent ones can’t attract users. |
| Contribution quality | 20/27 | 68 merged PRs out of 196 total submissions, 43 maintainer-closed unmerged PRs (39% rejection rate), 9 self-closed external PRs, 5 self-closed own-repo PRs; recent external PRs are almost entirely test refactors for langgenius/dify — you submit a lot, get bounced a lot, and most of your accepted work is low-impact cleanup on other people’s code. |
| Ecosystem / maintenance impact | 20/20 | 161 PRs + 45 commits across 23 popular repos (all-time, including 99 PRs and 41 commits to Dify’s 147k-star codebase; verified file samples confirm all are legitimate code changes, not spam). No extra adjustment was applied because the base score already priced that in. |
| Community influence | 3.4/8 | 8 followers, 8 following — perfect 1:1 follow ratio, zero organic social pull beyond the accounts you’ve chosen to follow yourself. |
| Activity authenticity | 15.7/17 | 600 contributions last year, active across 4 activity types, no bot-like or templated PR flooding detected — you’re actually writing code, not running a PR farm, even if most of it is low-impact test cleanup. |
Red flags
- 39% maintainer rejection rate on submitted PRs: you’ve worn the close button on other people’s repos thin with low-impact submissions.
- 83% of your recent external PRs target a single 147k-star repo (langgenius/dify), making your external credibility entirely dependent on one project’s maintainers’ tolerance for test refactors.
- 5 of your 10 original repos have zero stars, and your highest-starred original project only has 27 stars with no signs of meaningful user adoption.
Score calibration No extra adjustment. The base score already accounts for your legitimate external contributions, weak original project traction, and narrow PR focus.
Verdict Normal, low-risk contributor with a credible external track record, but you’re currently a test-refactor specialist for one big repo with a barren home portfolio — there’s room to build more unique, high-impact work before you can climb higher.