Other softwareAI models and assistantsLibraries and frameworksOperating systems and desktopMedia and creativeAI coding toolsInfrastructure and DevOpsTerminal and shellBrowsers and extensionsEmail and chatSelf-hosted and home serverNetworking and VPNEditors and IDEsSecurity and passwordsProductivityDatabasesFiles, sync and backupGit and code hostingNotes and knowledgeMonitoring and analyticsSearchFinance
AI coding tools
What HN commenters recommend, ranked by distinct recommenders. Under each: what people praise most and the most common caveat, or the strongest quote where there is no summary yet.
- 1.Loved for Daily reliability, Code review · Caveat Harness quality
- 2.Loved for Code quality, Reverse engineering · Caveat Security refusals
- 3.Loved for Broad automation, Output quality · Caveat Bugs
- 4.Loved for Extensibility, Lightweight minimalism · Caveat Plugin quality
- 5.Loved for Model flexibility, Cheap model access · Caveat Data privacy
- 6.Loved for Composer and Auto, Model choice · Caveat Pricing
- 7.Loved for IDE integration, Model choice · Caveat Pricing change
- 8.Loved for Daily driver, Speed · Caveat Low quotas
- 9.Loved for Batteries included, Multi-model roles · Caveat Context overhead
- 10.Loved for Structured workflow, Brainstorming skill · Caveat Token usage
- 11.Loved for Coding capability, Terminal interface · Caveat Behind competitors
- 12.Loved for Code quality, Model choice · Caveat Token cost
- 13.Loved for Terse responses, Token savings · Caveat Modest effect
- 14.Loved for Upfront planning, Clarifying ambiguity · Caveat Limited structure
- 15.Loved for Upfront planning, Constrains agents · Caveat Spec drift
- 16.Loved for Parallel worktrees, Daily driver · Caveat Cluttered UX
- 17.Loved for Agent continuity, Task planning · Caveat Bloat
- 18.Loved for Result quality, Handoff and threads · Caveat Price
- 19.Loved for Generous limits, Token value · Caveat Heavy-use limits
- 20.Loved for Open models, Provider flexibility · Caveat Defaults
- 21.Loved for Multi-harness support, Mobile experience · Caveat Not native CLI
- 22.Loved for Good OpenCode client, Agent features · Caveat OpenCode only
- 23.Loved for Model choice, Output quality · Caveat Ignores plan mode
- 24.Loved for Cheap model access, IDE integration · Caveat Model costs
- 25.Loved for TUI design, Task results · Caveat Codebase upload
- 26.Loved for Value for cost, Daily driver · Caveat Quality regressions
- 27.Loved for Remote hosts, Replaces alternatives · Caveat Hidden diff view
- 28.Loved for Speed, Architecture · Caveat Token usage
- 29.Loved for API docs lookup, Code quality
- 30.Loved for Works with others, Terse, small edits · Caveat Modest effect
- 31.Loved for Low cost, Cache hits · Caveat Data to China
- 32.Loved for Better UI output, Recommended
- 33.Loved for Productivity, Model flexibility · Caveat Unannounced changes
- 34.Loved for Emacs integration, Hackability · Caveat Claude breakage
- 35.Loved for Code quality, JetBrains integration · Caveat Behind competitors
- 36.Loved for Team-level agent, Sandboxed environment · Caveat Weak results
- 37.Loved for Fewer input tokens, Agent integration · Caveat Small savings
- 38.I recently changed to Maki and have been reasonably satisfied so far.
- 39.Even though we love their agent SDK, after claude agent SDK it feels like peanuts.
- 40.I use plannotator with claude a lot and have much better time
- 41.Antigravity CLI is quite decent, it's a huge step up from Gemini CLI
- 42.check out omnigent for building workflows across multiple harnesses
- 43.I’m a big fan of the Compound Engineering plugin from Every.
- 44.a good framework that follows "AI Agentic Agile Methodology", and specification driven development is BMAD. Really helpful for brainstorming sessions.
- 45.if you want an Actually Good Experience I recommend using Claude Code Router
- 46.This looks fantastic and I'm excited to try it today!
- 47.in the Claude Code realm I use GSD and it manages all that and keeps token use low.
- 48.Each time I start to implement a new ticket, superset will pull the ticket from linear, create a worktree/workspace, reserve ports, start the servers, start a browser and start Claude
- 49.The desktop app is actually really good (esp on macos).
- 50.Gas Town has a very clear "mad scientist/performance art" sort of thing going on, and I love that.
- 51.Windsurf - $20 gives u access to nearlly all models.
- 52.I recommend you look up https://github.com/github/spec-kit. It’s a project out of Microsoft/Github that encodes a rigorous spec-then-implement multi pass workflow.
- 53.I switched to Kiro and it's essentially identical.. well a bit better because you get a better idea of what the harness is doing
- 54.Goose was the only harness that didnt bloat context too much with system prompts
- 55.I love Hax. After a month on Codex, I switched to Hax.
- 56.It's great. Much stronger on the UX side for things that devs need than Claude / Codex.
- 57.Augment kind of became impossible to avoid once we realized the new guy is outpacing the rest of us
- 58.I now use Forge to plan css too because I don't want to show broken css states
- 59.I think relying on https://github.com/vercel-labs/skills is sound. It handles global installs for a wide range of coding agents.
- 60.I have been using jcode for past two weeks, honestly I feel its much better compared to Opencode.
- 61.Agent-of-empires seems pretty popular, has nice web and phone. Tmux based, which I dig.
- 62.I use serena as an MCP for all LSPs and it works great.
- 63.skills like wayfinder and grill-me (they are very helpful still)
- 64.I used Vercel's v0 to build a small business website for a relative in 10 minutes. It looked fantastic
- 65.Coding Plans by MiniMax ($20/mo for 1.7b tokens) and Z.ai (~$30/week use for $17/mo) are also tremendous value for money.
- 66.I recommend it if you're in search of something powerful with a web interface that works well on mobile devices
- 67.I use Claude Code Web w/ Opus / Fable to work on improvements on parallel branches
- 68.paired with CodeCompanion it is really powerful and ergonomic, way more them combining external harnesses
- 69.CodeNomad as the web-based GUI. It works perfectly; I can access it from anywhere
- 70.it's a thin wrapper around claude/codex/gemini, you can pick-up sessions from your phone and start terminal sessions remotely. It's pretty good.
- 71.I'm regularly extracting parts of my workflow to small, focused skills.
- 72.You should consider looking at oh-my-opencode for inspiration
- 73.
- 74.ZCode was surprisingly good, which is surprising for a tool that came out of nowhere
- 75.the reliability and almost absence of any error is amazing