REPOGEO 报告 · LITE
dirac-run/dirac
默认分支 master · commit c4f1df66 · 扫描时间 2026/5/28 08:28:17
星标 1,255 · Fork 68
行动计划告诉你下一步要做什么——按影响力排序、可直接复制粘贴的修改。品类可见性是真正的 GEO 测试:当用户向 AI 提一个不带品牌、本应让 dirac-run/dirac 浮出水面的问题时,AI 是真的推荐了你,还是推荐了你的竞品?客观检查验证 AI 引擎最先权衡的那些元数据信号。自指检查判断 AI 是否还认识你的名字。
行动计划 — 可复制粘贴的修复
3 条由 gemini-2.5-flash 生成、按优先级排序的修改。修完后请把对应条目标记为完成。
- highreadme#1Strengthen README's opening definition of the project
原因:
当前# Dirac - Accurate & Highly Token Efficient Open Source AI Agent > **Dirac topped the Terminal-Bench-2 leaderboard for `gemini-3-flash-preview` with a 65.2% score!** It is a well studied phenomenon that any given model's reasoning ability degrades with the context length. If we can keep context tightly curated, we improve both accuracy and cost while making larger changes tractable in a single task. Dirac is an open-source coding agent built with this in mind. It reduces API costs by **64.8%** on average while producing better and faster work. Using hash-anchored parallel edits, AST manipulation, and a suite of advanced optimizations. Oh, and no MCP.
复制粘贴的修复# Dirac - Accurate & Highly Token Efficient Open Source AI Agent Dirac is an open-source coding agent singularly focused on efficiency and context curation. It reduces API costs by 50-80% compared to other agents while simultaneously improving code quality, utilizing hash-anchored parallel edits, AST manipulation, and advanced optimizations. > **Dirac topped the Terminal-Bench-2 leaderboard for `gemini-3-flash-preview` with a 65.2% score!** It is a well studied phenomenon that any given model's reasoning ability degrades with the context length. If we can keep context tightly curated, we improve both accuracy and cost while making larger changes tractable in a single task.
- hightopics#2Add descriptive topics to improve categorization
原因:
复制粘贴的修复ai-agent, coding-assistant, llm-agent, code-refactoring, developer-tools, open-source, generative-ai, large-language-models
- mediumhomepage#3Set the repository homepage URL
原因:
复制粘贴的修复https://dirac.run/
本次扫描解析到的品类 GEO 通道:google/gemini-2.5-flash, deepseek/deepseek-v4-flash
品类可见性 — 真正的 GEO 测试
向 google/gemini-2.5-flash 提出的不带品牌问题。AI 推荐了你,还是推荐了别人?
各模型使用同一组问题 — 切换标签对比回答与排名。
- GitHub Copilot Enterprise · 被推荐 1 次
- Tabnine Enterprise · 被推荐 1 次
- Codeium Enterprise · 被推荐 1 次
- JetBrains AI Assistant · 被推荐 1 次
- OpenAI API · 被推荐 1 次
- 品类问题Looking for an AI coding assistant that reduces API costs and improves code quality.你:未被推荐AI 推荐顺序:
- GitHub Copilot Enterprise
- Tabnine Enterprise
- Codeium Enterprise
- JetBrains AI Assistant
- OpenAI API
AI 推荐了 5 个替代方案,却始终没点名 dirac-run/dirac。这就是要补上的差距。
查看 AI 完整回答
- 品类问题What open-source AI agents are best for efficient code refactoring and context management?你:未被推荐AI 推荐顺序:
- Cursor
- Continue (continue-dev/continue)
- OpenDevin (OpenDevin/OpenDevin)
- GPT-Engineer (gpt-engineer-org/gpt-engineer)
- Auto-GPT (Significant-Gravitas/Auto-GPT)
- AgentGPT (reworkd/AgentGPT)
- SuperAGI (TransformerOptimus/SuperAGI)
- Ollama (ollama/ollama)
AI 推荐了 8 个替代方案,却始终没点名 dirac-run/dirac。这就是要补上的差距。
查看 AI 完整回答
客观检查
针对 AI 引擎最看重的元数据信号的规则审计。
- Metadata completenesswarn
建议:
- README presencepass
自指检查
当被直接问到你时,AI 是否还知道你的仓库存在?
- Compared to common alternatives in this category, what is the core differentiator of dirac-run/dirac?passAI 明确点名了 dirac-run/dirac
AI 的回答可能信誓旦旦却是错的。请按事实核对:技术栈、目标人群、差异化点是不是和你实际的对得上?
- If a team adopts dirac-run/dirac in production, what risks or prerequisites should they evaluate first?passAI 明确点名了 dirac-run/dirac
AI 的回答可能信誓旦旦却是错的。请按事实核对:技术栈、目标人群、差异化点是不是和你实际的对得上?
- In one sentence, what problem does the repo dirac-run/dirac solve, and who is the primary audience?passAI 明确点名了 dirac-run/dirac
AI 的回答可能信誓旦旦却是错的。请按事实核对:技术栈、目标人群、差异化点是不是和你实际的对得上?
嵌入你的 GEO 徽章
把这个徽章贴进 dirac-run/dirac 的 README。每次重新扫描都会自动更新,并跳到最新报告——是「我在乎 AI 可发现性」最简单的公开证明。
[](https://repogeo.com/zh/r/dirac-run/dirac)<a href="https://repogeo.com/zh/r/dirac-run/dirac"><img src="https://repogeo.com/badge/dirac-run/dirac.svg" alt="RepoGEO" /></a>订阅 Pro,解锁深度诊断
dirac-run/dirac — 轻量扫描仍免费;本卡列出 Pro 相对轻量的深度额度。
- 深度报告每月 10 次
- 无品牌品类查询5,轻量 2
- 优先行动项8,轻量 3