[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"pack-detail-aider-pair-programming-es":3,"seo:pack:aider-pair-programming:es":109},{"code":4,"message":5,"data":6},200,"操作成功",{"pack":7},{"slug":8,"icon":9,"tone":10,"status":11,"status_label":12,"title":13,"description":14,"items":15,"install_cmd":108},"aider-pair-programming","🤝","#FF6347","new","Nuevo · esta semana","Esenciales de Programación en Pareja con Aider","Diez picks para el dev que quiere pair programming nativo en terminal con un commit git por edición. Aider core + config + watch + convenciones + LiteLLM + modelo local + herramientas git + cuándo cambiar de agente. En orden de instalación.",[16,28,37,47,57,67,74,84,92,100],{"id":17,"uuid":18,"slug":19,"title":20,"description":21,"author_name":22,"view_count":23,"vote_count":24,"lang_type":25,"type":26,"type_label":27},2786,"96b9add2-d3ab-466a-817b-0ec2c462378b","aider-ai-pair-programming-in-your-terminal","Aider — AI Pair Programming in Your Terminal","Aider is the OG terminal AI pair programmer. Edits multiple files in one prompt, auto-commits each change. 36K+ stars, BYOK 100+ providers via LiteLLM.","Aider",24,0,"en","script","Script",{"id":29,"uuid":30,"slug":31,"title":32,"description":33,"author_name":22,"view_count":34,"vote_count":24,"lang_type":25,"type":35,"type_label":36},2787,"553b65cc-501c-4012-97d9-ed08ca3c6415","aider-conf-yml-production-aider-configuration",".aider.conf.yml — Production Aider Configuration","Drop-in `.aider.conf.yml` template with project Aider settings — auto-test, lint, model split, file ignores. Production-tested from 36K-star community.",23,"config","Config",{"id":38,"uuid":39,"slug":40,"title":41,"description":42,"author_name":43,"view_count":44,"vote_count":24,"lang_type":25,"type":45,"type_label":46},638,"e0565f2a-21a4-4f82-8ba2-1f77ac970144","awesome-claude-md-project-config-best-practices-e0565f2a","Awesome CLAUDE.md — Project Config Best Practices","Community collection of CLAUDE.md templates and best practices for configuring Claude Code per-project. Includes templates for monorepos, web apps, Python projects, and team workflows. 3,000+ stars.","Prompt Lab",115,"prompt","Prompt",{"id":48,"uuid":49,"slug":50,"title":51,"description":52,"author_name":53,"view_count":54,"vote_count":24,"lang_type":25,"type":55,"type_label":56},2789,"0f113965-1adc-4435-982b-fb613fa4d157","litellm-proxy-unified-gateway-for-100-llm-apis","LiteLLM Proxy — Unified Gateway for 100+ LLM APIs","LiteLLM Proxy maps 100+ LLM providers (Anthropic, OpenAI, Bedrock, Vertex) to one OpenAI-compatible endpoint. Auth, rate limit, cost track, fallbacks.","LiteLLM (BerriAI)",81,"agent","Agent",{"id":58,"uuid":59,"slug":60,"title":61,"description":62,"author_name":63,"view_count":64,"vote_count":24,"lang_type":25,"type":65,"type_label":66},2834,"08acf3a7-b56b-40d2-9c94-9a8eb773eca4","deepseek-coder-code-specialized-model-for-local-inference","DeepSeek Coder — Code-Specialized Model for Local Inference","DeepSeek Coder is the code-specialized open-weight model with FIM (fill-in-middle) support. Beats Codestral on HumanEval. Drops into Continue, Aider.","DeepSeek",62,"skill","Skill",{"id":68,"uuid":69,"slug":70,"title":71,"description":72,"author_name":22,"view_count":73,"vote_count":24,"lang_type":25,"type":55,"type_label":56},2788,"de82f878-b0f0-4dfe-ae99-d4fae2ae76e8","aider-watch-mode-auto-edit-on-ai-comments","Aider Watch Mode — Auto-Edit on `# AI:` Comments","Aider Watch Mode tails files and edits when you save an `# AI:` comment. Trigger Aider from VS Code, vim, Cursor, Claude Code without terminal switch.",14,{"id":75,"uuid":76,"slug":77,"title":78,"description":79,"author_name":80,"view_count":81,"vote_count":24,"lang_type":25,"type":82,"type_label":83},803,"85eacf96-3b2b-43f5-ad6f-fd4413ad0b69","git-mcp-version-control-server-ai-agents-85eacf96","Git MCP — Version Control Server for AI Agents","MCP server that gives AI agents full Git capabilities. Git MCP enables Claude Code and Cursor to manage repositories, branches, commits, diffs, and history via tool calls.","MCP Hub",180,"mcp","MCP",{"id":85,"uuid":86,"slug":87,"title":88,"description":89,"author_name":90,"view_count":91,"vote_count":24,"lang_type":25,"type":65,"type_label":66},1036,"6ed075de-35cb-11f1-9bc6-00163e2b0d79","lazygit-simple-terminal-ui-git-commands-6ed075de","Lazygit — Simple Terminal UI for Git Commands","Lazygit is a simple terminal UI for git commands. Stage files, commit, push, rebase, cherry-pick, squash, and browse history without memorizing 30 git commands. Written in Go and designed for speed with vim-style keybindings.","Script Depot",188,{"id":93,"uuid":94,"slug":95,"title":96,"description":97,"author_name":98,"view_count":99,"vote_count":24,"lang_type":25,"type":65,"type_label":66},1037,"6ed07802-35cb-11f1-9bc6-00163e2b0d79","tmux-terminal-multiplexer-sessions-windows-panes-6ed07802","tmux — Terminal Multiplexer for Sessions, Windows, Panes","tmux is a terminal multiplexer that lets you switch between several programs in one terminal, detach them, and reattach them to a different terminal. Essential for SSH workflows, long-running processes, and pair programming.","AI Open Source",97,{"id":101,"uuid":102,"slug":103,"title":104,"description":105,"author_name":106,"view_count":107,"vote_count":24,"lang_type":25,"type":45,"type_label":46},841,"b3a901fe-8380-4bfc-a631-c57f8d5b51f0","ai-coding-agent-comparison-2026-complete-guide-b3a901fe","AI Coding Agent Comparison 2026 — Complete Guide","Comprehensive comparison of all AI coding agents in 2026. Covers Claude Code, Cursor, Codex, Gemini CLI, Cline, Roo Code, Windsurf, and Aider with feature matrices and recommendations.","Agent Toolkit",226,"tokrepo install pack\u002Faider-pair-programming",{"pageType":110,"pageKey":8,"locale":25,"title":111,"metaDescription":112,"h1":113,"tldr":114,"bodyMarkdown":115,"faq":116,"schema":132,"internalLinks":138,"citations":151,"wordCount":164,"generatedAt":165},"pack","Aider Pair Programming Essentials — 10 Picks for Terminal-Native AI Coding","Aider is the OG terminal AI pair programmer with a git commit per edit. Ten opinionated picks: install, configure, add watch mode + project conventions, route through LiteLLM, drop in DeepSeek for cheap local turns, add Git MCP + Lazygit + tmux, and know when to switch to Claude Code or Codex. Install in order.","Aider Pair Programming Essentials — Set Up the OG Terminal Coding Agent","Ten picks in install order: Aider itself, .aider.conf.yml, AGENTS.md\u002FCONVENTIONS.md style project guidelines, LiteLLM proxy so you can swap models without rewriting calls, DeepSeek Coder for cheap turns, watch mode so any editor triggers Aider, Git MCP and Lazygit to actually use the commit-per-edit trail, tmux to run tests alongside, and the coding agent comparison so you know when Aider is wrong for the job. Skip the YouTube rabbit hole — these ten get you productive.","## What's in this pack\n\nAider is the OG. Before Claude Code, before Codex CLI, before Cursor's agent mode — Paul Gauthier shipped `aider` and proved you could pair-program with an LLM from a terminal, with a real git commit per edit. It still has the cleanest commit log of any agent and the widest model coverage (anything LiteLLM supports, which is 100+ providers).\n\nThis pack is for the developer who has heard about Aider, wants to try it seriously, and doesn't want to spend Saturday watching install videos. **Ten picks, in install order**, that get you from `pip install aider-chat` to a real workflow — config locked in, project guidelines auto-loaded, a cheap fallback model, watch mode so your editor of choice can trigger it, git tooling to actually review the commits, and the comparison guide so you know when to reach for Claude Code or Codex instead.\n\nWho this is for: a terminal-comfortable developer working in a real codebase who values **explicit edits + per-change git commits** over agentic exploration. If you want autonomous multi-step planning, this is the wrong stack — go install Claude Code. If you want \"AI as a precise scalpel that records every cut,\" keep reading.\n\n## Install in this order\n\n1. **Aider — AI Pair Programming in Your Terminal** — `pip install aider-chat`, point at an API key, run `aider file1.py file2.py`. That's it. The whole experience is: name the files you want to edit, describe the change, watch Aider produce a diff and commit it. Five minutes to feel the loop; the rest of this pack is making it stop being clunky.\n2. **.aider.conf.yml — Production Aider Configuration** — Drop this template in your project root. It pins your default model, your weak model for cheap turns, lint and test commands Aider runs after each edit, file ignores, and architect\u002Feditor model split. The single highest-leverage file after Aider itself. Without it you'll fight the defaults forever.\n3. **Awesome CLAUDE.md — Project Config Best Practices** — Aider reads `CONVENTIONS.md` (or `AGENTS.md`, or whatever you point `--read` at) the same way Claude Code reads CLAUDE.md. The patterns in this collection — terse stack notes, hard NOs, naming rules — transfer directly. Pick one template, rename it `CONVENTIONS.md`, add it to `.aider.conf.yml` under `read:`.\n4. **LiteLLM Proxy — Unified Gateway for 100+ LLM APIs** — Aider already speaks LiteLLM natively, but running the proxy gives you per-project keys, spend caps, and a unified log of every call. The proxy also lets you switch between Anthropic, OpenAI, OpenRouter, and local Ollama without touching `.aider.conf.yml`. Set this up before your first $300 month.\n5. **DeepSeek Coder — Code-Specialized Model for Local Inference** — Aider's `--weak-model` is the cheap model it uses for commit messages and routine edits. DeepSeek Coder is the canonical cheap-and-good choice — runs locally via Ollama or through DeepSeek's API at a fraction of GPT-4 pricing. Big quality jump for negligible cost.\n6. **Aider Watch Mode — Auto-Edit on `# AI:` Comments** — Run `aider --watch-files` and now any editor (VS Code, vim, Cursor, Claude Code, JetBrains) can trigger Aider just by saving a file containing `# AI:` comments. This bridges the \"I want to stay in my IDE\" gap. Aider runs in a terminal tab; you never leave your editor.\n7. **Git MCP — Version Control Server for AI Agents** — Aider commits per edit; that's a feature. But the commit *history* becomes useful only when other tools can read it. Git MCP exposes the local repo to any MCP-aware agent (Claude Code, Cursor, Gemini CLI) so they can see exactly what Aider just did and reason about it. Pairs perfectly with the multi-agent setups in pick #10.\n8. **Lazygit — Simple Terminal UI for Git Commands** — Aider's commit-per-edit is great in theory; in practice you'll want to squash 12 micro-commits into 1 PR-worthy commit before pushing. Lazygit's interactive rebase is the fastest way to do that without leaving terminal. Hit `R` on the base commit, mark commits as `squash`, write a clean message. 20 seconds vs 2 minutes of `git rebase -i` flag-juggling.\n9. **tmux — Terminal Multiplexer for Sessions, Windows, Panes** — Aider in one pane, your test runner watching files in another, server logs in a third. The whole Aider experience improves 3x once you stop alt-tabbing. Detach the session, come back tomorrow, your model context is gone but Aider re-reads the files and you're back where you were.\n10. **AI Coding Agent Comparison 2026 — Complete Guide** — Read this when Aider feels wrong. It will sometimes — multi-file refactors, exploratory work where you don't know which files to add, anything requiring tool use beyond shell. This guide tells you when to switch to Claude Code (agentic exploration), Codex CLI (autonomous tasks), or Cursor (IDE-integrated). Aider is a precision tool, not a swiss army knife.\n\n## How they fit together\n\n```\nAider (#1)\n   │\n   └─ the edit loop: name files → describe → diff → commit\n         │\n.aider.conf.yml (#2)\n   │\n   ├─ pins model + weak model\n   ├─ runs lint + test after edits\n   └─ loads CONVENTIONS.md\n         │\nCONVENTIONS.md via Awesome CLAUDE.md (#3)\n   │\n   └─ project memory Aider reads every session\n         │\nLiteLLM Proxy (#4)\n   │\n   ├─ unified model gateway\n   └─ spend caps per project\n         │\nDeepSeek Coder (#5)\n   │\n   └─ cheap weak-model for commit msgs + routine edits\n         │\nWatch Mode (#6)\n   │\n   └─ any editor → save # AI: comment → Aider edits\n         │\nGit MCP (#7) + Lazygit (#8)\n   │\n   ├─ Git MCP: other agents can read Aider's trail\n   └─ Lazygit: squash micro-commits before push\n         │\ntmux (#9)\n   │\n   └─ Aider + tests + logs in one window\n         │\nComparison Guide (#10) — know when to reach for a different tool.\n```\n\nThe load-bearing trio is **Aider + .aider.conf.yml + CONVENTIONS.md**. Everything else is upgrade. If you only have 15 minutes: install #1, drop in #2, point it at a small `CONVENTIONS.md` and run.\n\n## Tradeoffs you'll hit — Aider vs Claude Code vs Codex CLI\n\n- **Aider vs Claude Code** — Aider is *explicit*: you say \"edit these 3 files\" and it edits those 3 files. Claude Code is *agentic*: you describe the task and it decides which files to touch, runs tools, iterates. Aider wins on precision and git hygiene (every edit is a real commit you can revert). Claude Code wins on exploratory work and multi-step tasks. Use Aider for surgical changes you can describe in one sentence; use Claude Code when you'd otherwise spend 20 minutes finding the right files.\n- **Aider vs Codex CLI** — Codex CLI is OpenAI's terminal agent — closer to Claude Code in mental model (autonomous, agentic). Aider is fundamentally different: human-driven, file-scoped, commit-per-edit. If you want OpenAI models with Aider's discipline, you can route Aider through GPT-4o via LiteLLM and get most of the Codex experience plus the commit log. If you want Codex's autonomy, Aider won't match it.\n- **Model split (architect vs editor)** — Aider's killer feature most people skip: set `architect_model: o1` (or Claude Opus) for planning, `editor_model: claude-sonnet` (or DeepSeek Coder) for the actual diff. Architect plans cheap and slow; editor edits fast and cheap. Quality goes up, cost goes down. The `.aider.conf.yml` template (#2) has this preconfigured.\n- **Commit-per-edit vs feature commits** — Aider's default of one commit per edit feels noisy if you're used to crafting PR-worthy commits. Two options: (a) Lazygit (#8) squash before push; (b) `--no-auto-commit` and commit manually. Most people stick with auto-commit + squash because the per-edit history is invaluable when something breaks and you need to `git bisect` your AI sessions.\n\n## Common pitfalls\n\n- **Burning $50 in one session** — Aider sends the whole file (or the whole repo map) on every turn. Without `.aider.conf.yml` setting a `weak_model`, every commit-message generation uses your expensive model. Set it up before you start. Watch `\u002Ftokens` inside Aider — it tells you the cost of the current context.\n- **Edit-mode confusion (whole vs diff vs udiff vs architect)** — Aider has multiple edit formats and picks one per model. If edits are failing (\"diff not applied\"), check `--edit-format`. Sonnet does best with `diff-fenced`; o1 prefers `architect` mode where it plans then a cheaper editor model writes the diff. Read the model recommendations once; never think about it again.\n- **Auto-commits filling history with garbage** — Yes, by default Aider commits everything. The fix isn't `--no-auto-commit`; it's discipline at PR time. Use Lazygit (#8) to squash the session into a single conventional commit. Or set up a pre-push hook that warns if you have more than N un-squashed AI commits.\n- **Forgetting `\u002Fadd` and `\u002Fdrop`** — Aider only edits files you've added. Adding the whole repo blows context; adding too few files means edits skip files that need touching. Use `\u002Fadd file.py` and `\u002Fdrop file.py` liberally. The repo map (auto-included by default) is usually enough context for Aider to know *what exists* without you adding everything.\n- **Treating watch mode as a Cursor replacement** — It's not. Cursor inlines suggestions in the editor; Aider watch mode requires you to save a `# AI:` comment and wait for a diff. Different mental model. If you want \"AI types as you type,\" use Cursor or Copilot. If you want \"I'll describe what I want and you do it precisely,\" Aider watch mode is the workflow.",[117,120,123,126,129],{"q":118,"a":119},"Why Aider when Claude Code and Cursor exist?","Three reasons. (1) Git discipline: every edit is a real commit. When something breaks, you can bisect; you can revert a single sentence-sized change. No other agent does this as cleanly. (2) Model coverage: anything LiteLLM supports (100+ providers, including local Ollama) works in Aider — Claude Code is Anthropic-locked, Cursor is somewhat model-locked. (3) Explicit scope: you tell Aider which files to touch, it touches only those. Predictable in a way that agentic agents aren't. Aider is the right tool when you know exactly what you want changed and want a clean audit trail; it's the wrong tool for 'go explore the codebase and figure it out.'",{"q":121,"a":122},"Will I burn through tokens with auto-commit on every edit?","Auto-commit itself is local — it costs zero tokens (just `git commit`). What costs tokens is the commit *message* generation, which by default uses your main model. Fix: set `weak_model: deepseek-coder` (or any cheap model) in `.aider.conf.yml` and Aider routes commit messages there. Combined with `architect_model` for planning + `editor_model` for diffs, you can run Aider all day for the price of a coffee. The first month without these splits is everyone's wake-up call.",{"q":124,"a":125},"Does watch mode (#6) work with VS Code, Cursor, or Claude Code?","Yes — it's editor-agnostic. Watch mode tails files on disk. Save any file containing an `# AI: \u003Cinstruction>` comment (or `\u002F\u002F AI:` for JS\u002FGo, `# AI?` for questions) and Aider picks it up. It works in vim, VS Code, Cursor, Claude Code's edit mode, JetBrains, even nano. The only requirement is your editor writes to disk on save (most do). You can even run Aider in a tmux pane and use Claude Code in another — they coexist over the same git tree fine, especially with Git MCP (#7) giving Claude visibility into Aider's commits.",{"q":127,"a":128},"Why include LiteLLM Proxy (#4) when Aider already supports LiteLLM directly?","Aider talks to LiteLLM as a Python library, which is enough for one user. The *proxy* server adds three things Aider alone doesn't: (a) shared rate limits and spend caps across multiple tools (Aider + Claude Code + Cursor all going through one budget), (b) a unified log of every LLM call across your whole stack — invaluable for debugging cost spikes, (c) per-project API keys you can revoke. If you're a solo dev using only Aider, you can skip the proxy. If you have any other AI tool installed, the proxy pays for itself in week one.",{"q":130,"a":131},"Aider crashes \u002F loses context when files are huge. Workarounds?","Three plays. (1) `\u002Fdrop` unused files mid-session — Aider keeps full content of added files in context every turn, so dropping a 2000-line file you no longer need cuts cost and improves quality. (2) Use the repo map instead of `\u002Fadd`-ing everything — Aider auto-builds a tree-sitter symbol map of the whole repo and feeds Claude\u002FGPT the relevant chunks; you only `\u002Fadd` the files actually being edited. (3) Architect mode (`architect_model: o1`) — o1 plans the change as text, then a cheaper editor model writes the diff over only the files mentioned in the plan. This is the cleanest way to handle 'edit this large refactor across 8 files' without blowing context.",{"@context":133,"@type":134,"name":135,"description":136,"numberOfItems":137,"inLanguage":25},"https:\u002F\u002Fschema.org","ItemList","Aider Pair Programming Essentials","Ten opinionated picks for setting up Aider — the OG terminal AI pair programmer with a git commit per edit. In install order.",10,[139,143,147],{"url":140,"anchor":141,"reason":142},"\u002Fen\u002Ftopics","Browse other topic packs","Adjacent packs cover Claude Code, Codex CLI, and the broader terminal-agent stack",{"url":144,"anchor":145,"reason":146},"\u002Fen\u002Fai-tools-for\u002Fcoding","AI tools for coding","Compare Aider against the full landscape of AI coding agents in 2026",{"url":148,"anchor":149,"reason":150},"\u002Fen\u002Ffeatured","Featured assets on TokRepo","Once Aider is set up, browse the wider curated catalog for adjacent tools",[152,156,160],{"claim":153,"source_name":154,"source_url":155},"Aider supports 100+ LLM providers via LiteLLM and commits each edit to git","Aider official docs","https:\u002F\u002Faider.chat",{"claim":157,"source_name":158,"source_url":159},"Aider watch mode tails files and triggers on `# AI:` comments from any editor","aider.chat — Edit files with your editor","https:\u002F\u002Faider.chat\u002Fdocs\u002Fusage\u002Fwatch.html",{"claim":161,"source_name":162,"source_url":163},"LiteLLM provides a unified proxy for 100+ LLM providers with spend caps and unified logging","BerriAI\u002Flitellm on GitHub","https:\u002F\u002Fgithub.com\u002FBerriAI\u002Flitellm",1555,"2026-05-22T00:00:00Z"]