惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

WordPress大学
WordPress大学
The Cloudflare Blog
大猫的无限游戏
大猫的无限游戏
小众软件
小众软件
V
Visual Studio Blog
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
博客园 - 司徒正美
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
酷 壳 – CoolShell
酷 壳 – CoolShell
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
月光博客
月光博客
The GitHub Blog
The GitHub Blog
L
LangChain Blog
D
DataBreaches.Net
T
The Blog of Author Tim Ferriss
F
Fortinet All Blogs
博客园 - Franky
阮一峰的网络日志
阮一峰的网络日志
GbyAI
GbyAI
Apple Machine Learning Research
Apple Machine Learning Research
宝玉的分享
宝玉的分享
Engineering at Meta
Engineering at Meta
V
V2EX
MyScale Blog
MyScale Blog

Hacker News - Newest: "AI"

AI can't read an investor deck AI as an attorney? Student uses ChatGPT, Gemini to sue UW over alleged racial discrimination Hacking MCP Servers in AI Systems – The Rug Pull: Tool Changes After Approval GitHub - MeepCastana/KubeezCut: Free Web based video editor Can AI judge journalism? A Thiel-backed startup says yes, even if it risks chilling whistleblowers Coming soon: 10 Things That Matter in AI Right Now DARPA built an AI to fact-check enemy weapons claims What explains heterogeneity in AI adoption? When AI Meets Muscle: Context-Aware Electrical Stimulation Promises a New Way to Guide Human Movements - Department of Computer Science AI Changed How We Build. It Did Not Change What Matters. Linux rules on using AI-generated code - Copilot is OK, but humans must take 'full responsibility for the… Meta spins up AI version of Mark Zuckerberg to engage with employees Code Mode: Let Your AI Write Programs, Not Just Call Tools | TanStack Blog GitHub - Delavalom/graft: Go framework for building AI agents. Type-safe tools, multi-provider (OpenAI, Anthropic, Gemini, Bedrock), zero vendor SDKs. India's TCS tops estimates, says new AI models did not dent services demand Gen Z's fading AI hype Strong feeling: we are in a folded AI reality GitHub - machinarii/total-recall-catalog: A reference catalog of latest knowledge retrieval, memory & RAG systems GitHub - mensfeld/code-on-incus: Give each AI agent its own isolated machine with root, Docker, and systemd. Active defense detects and stops threats automatically.. Quantization, LoRA, and the 8% Problem: Benchmarking Local LLMs for Production AI Iran war: We spoke to the man making Lego-style AI videos that experts say are powerful propaganda Powell, Bessent discussed Anthropic's Mythos AI cyber threat with major U.S. banks GitHub - immartian/bellamem: Persistent belief-graph memory for AI agents. Retrieves decisive context by importance — not recency, not RAG, not /compact. recursive-mode: The Repo-Native Operating System for AI Engineering After the attack on Sam Altman's home, will AI CEO's go on the offensive? The biggest advance in AI since the LLM Opus 4.6 vs GPT 5.4 One Prompt Unity World Generation Test “AI polls” are fake polls Client Challenge Can AI be a 'child of God'? Inside Anthropic's meeting with Christian leaders
GitHub - cachebag/jumpy: Sweet-spot between trad programm...
cachebag · 2026-05-18 · via Hacker News - Newest: "AI"

Inspired by 99: I, like Prime, wanted a tool to allow for me to still be in the driver seat if I am going to spawn an AI to edit my code.


image image

The main difference between jumpy and other tools, like 99, are:

feature jumpy avante / sidekick codecompanion inline claude code / aider
Interaction model Inline prompt → hunks in buffer Sidebar chat → apply Inline edit → accept/reject CLI agent → full file writes
Review granularity Per-hunk accept/reject Per-file or per-suggestion Per-change Post-hoc (git diff)
Context switch None: you stay in buffer Sidebar opens Minimal Leave editor or split terminal
LLM output format Search/replace blocks Full file / patch Full file Full file via temp
Token efficiency High (only changed lines sent back) Lower (full context) Lower Lower
Scope Single targeted edit Multi-file refactors, chat Single edit or chat Whole-project agentic tasks

Some of these are to be fleshed out, but the sole purpose is to know exactly what I am letting the LLM write into my code. I have no interest in letting it change everything, and only then can I go back and review every change.

So the philosophy here is: yes, I still want to handwrite my code AND use AI, but I prefer that I have full control over what the AI is writing. I don't want to have to wait until it finishes in order to review its changes.

That being said, I don't know wtf I'm doing, and right now, this really doesn't work very well.

But hopefully it will soon :P

Install

-- lazy.nvim
{
  "cachebag/jumpy",
  config = function()
    require("jumpy").setup({
      provider = "anthropic", -- or "openai", "openrouter"
    })
  end,
}

set your API key: export ANTHROPIC_API_KEY="sk-ant-..." (or OPENAI_API_KEY, JUMPY_API_KEY for openrouter).

Use

Keybind Action
<leader>j Open prompt, type your change, hit <CR>
]h / [h Next / previous hunk
<leader>a Accept hunk
<leader>x Reject hunk
<leader>A Accept all hunks
<leader>X Reject all hunks
<leader>r Reprompt the hunk under cursor

License

MIT