{"slug": "superpowers", "name": "Superpowers", "category": "ai-coding", "category_name": "AI coding tools", "description": "Open-source collection of agent skills and workflows for Claude Code and other coding agents.", "recommenders": 22, "commenters": 46, "warners": 8, "score": 17.069, "rising": 3.42, "first": "2025-11-28", "latest": "2026-10-19", "why": "Recommended as a skills plugin that structures coding-agent work into brainstorming, planning and review steps, with the /brainstorming skill singled out most often.", "why_is_quote": false, "has_verdict": true, "sentiment": {"recommend": 22, "neutral": 13, "compare": 3, "warn": 8}, "spark": [0, 1, 1, 1, 2, 2, 1, 2, 4, 1, 4, 3], "listed": true, "loved": ["Structured workflow", "Brainstorming skill"], "caveat": "Token usage", "short_quote": "Superpowers is one of the few skill/agent suites I use for all software development projects.", "icon": true, "domain": null, "github_repo": "github.com/obra/superpowers", "official_url": "https://github.com/obra/superpowers", "pricing": "open source", "aliases": ["Superpowers", "superpowers"], "verdict": [{"text": "Recommended as a skills plugin that structures coding-agent work into brainstorming, planning and review steps, ", "cites": [{"id": 797687409, "item_id": 49855463, "url": "https://news.ycombinator.com/item?id=49855463", "by": "parasti", "date": "2026-09-26", "stance": "recommend", "quote": "superpowers applies a fairly high level of scrutiny at every step. It's really quite effective", "excerpt": "At work, we use Claude Code with superpowers which organizes most work into planning followed by review followed by implementation and superpowers applies a fairly high level of scrutiny at every step. It's really quite effective, I don't generally read the code that comes out. In that workflow I have never used plan mode. I don't know what I'd use it for. On my hobby projects I use Antigravity CLI and almost always ", "context": "", "use_case": "planning, review, implementation workflow", "switched_from": "", "story": "Plan mode is dead"}, {"id": 799794784, "item_id": 49987174, "url": "https://news.ycombinator.com/item?id=49987174", "by": "Merad", "date": "2026-10-07", "stance": "recommend", "quote": "Before I discovered superpowers I was manually telling claude to do these workflows, the plugin makes it easy.", "excerpt": "This repo I started yesterday has most of the patterns I like to use: https://github.com/mbcrawfo/SqlSource Really the single biggest piece I use today is using a tool like superpowers to ensure your task is well defined. Added bonuses are its TDD implementation skills and multiple review steps baked into the workflow. Before I discovered superpowers I was manually telling claude to do these workflows, the plugin mak", "context": "", "use_case": "defining tasks, TDD workflow", "switched_from": "", "story": "JetBrains reports revenue growth, net financial loss for 2025"}, {"id": 740226464, "item_id": 46264154, "url": "https://news.ycombinator.com/item?id=46264154", "by": "radio879", "date": "2025-12-14", "stance": "recommend", "quote": "I\u2019ve been using the Superpowers plan mode: it creates a detailed plan first, then there\u2019s an execute mode", "excerpt": "Lately, I\u2019ve been using Claude Code skills/plugins. One is called Superpowers (you should be able to find it by typing \u201cclaude code superpowers github\u201d), and there are many other skills installed for all sorts of things. Gotta give props to whoever came up with the skills idea at Anthropic. It\u2019s a smart way of handling context \u2014 instead of dumping pages of helper text into the prompt (for example, how to properly bui", "context": "", "use_case": "planning and executing long coding tasks", "switched_from": "", "story": "Ask HN: How can I get better at using AI for programming?"}]}, {"text": "with the /brainstorming skill singled out most often.", "cites": [{"id": 797710576, "item_id": 49856911, "url": "https://news.ycombinator.com/item?id=49856911", "by": "teliskr", "date": "2026-09-26", "stance": "recommend", "quote": "I've been using superpowers:brainstorming skill with claude for a while and have found the process to work well for me.", "excerpt": "I've been using superpowers:brainstorming skill with claude for a while and have found the process to work well for me. I generally spend 10-30 minutes typing my requirements and ideas for a feature and then I feed it into brainstorming skill which asks questions to fill in the gaps before starting to build things. I don't bother reading the spec it creates. I have been consistently happy with the results. However, s", "context": "", "use_case": "brainstorming feature specs", "switched_from": "", "story": "Plan mode is dead"}, {"id": 776135120, "item_id": 48508445, "url": "https://news.ycombinator.com/item?id=48508445", "by": "spgorbatiuk", "date": "2026-06-12", "stance": "recommend", "quote": "I really recommend the /brainstorming skill there.", "excerpt": "In an unlikely scenario you haven't stumbled upon the superpowers (https://claude.com/plugins/superpowers), I really recommend the /brainstorming skill there. The plugin itself leans into engineering domain, but this specific skill makes claude think better and bring up and ask thoughtful stuff before it gets to a solution part", "context": "", "use_case": "improving Claude's reasoning", "switched_from": "", "story": "Ask HN: Favorite prompts for improving LLM output?"}, {"id": 782438033, "item_id": 48902377, "url": "https://news.ycombinator.com/item?id=48902377", "by": "ksaun", "date": "2026-07-13", "stance": "recommend", "quote": "For any significant new feature or work, I use /brainstorming (superpowers plug-in).", "excerpt": "Overall, I've been very pleased, though I have seen inconsistency like you mention. First, some background: At the start, I used Sonnet, but after a couple weeks, I'd switched to mostly Opus. I've used Fable when available. For any significant new feature or work, I use /brainstorming (superpowers plug-in). For Vestiges, I haven't written any design documentation per se. I'll start off a /brainstorming with a hefty p", "context": "building a game called Vestiges", "use_case": "feature brainstorming and design", "switched_from": "", "story": "Ask HN: What Are You Working On? (July 2026)"}]}, {"text": " The most common complaint is heavy token usage, ", "cites": [{"id": 802445664, "item_id": 50152854, "url": "https://news.ycombinator.com/item?id=50152854", "by": "moondowner", "date": "2026-10-19", "stance": "recommend", "quote": "I've been using Superpowers for months, it's great, only downside is the extensive token usage.", "excerpt": "I've been using Superpowers for months, it's great, only downside is the extensive token usage. When I saw the title Superlite, I thought it's a Superpowers fork optimised for less token usage...", "context": "", "use_case": "AI coding workflow", "switched_from": "", "story": "Superlite: A Faster Alternatie to Superpowers"}, {"id": 800237824, "item_id": 50014864, "url": "https://news.ycombinator.com/item?id=50014864", "by": "tombert", "date": "2026-10-09", "stance": "warn", "quote": "I don't feel like it gave me significantly better results than just raw-dogging it.  It did however burn through my tokens significantly faster.", "excerpt": "I used Superpowers for awhile, but I don't feel like it gave me significantly better results than just raw-dogging it. It did however burn through my tokens significantly faster. Maybe I'll come to miss it now that I removed it, but I certainly don't yet.", "context": "", "use_case": "AI coding skills", "switched_from": "", "story": "Why isn't the industry freaking out about DeepSeek 4.1 Flash?"}, {"id": 799483088, "item_id": 49967693, "url": "https://news.ycombinator.com/item?id=49967693", "by": "sajithdilshan", "date": "2026-10-05", "stance": "warn", "quote": "It ended up burning so much tokens and taking actually longer to finish and just could have been done without any agent. I immediately uninstall the plugin.", "excerpt": "It's not necessarily corporate, but mainly due to the fact that AI tools are quite new and we're just learning to use them efficiently. One example is that recently I used the superpowers claude code plugin. I thought it was good because it was popular on github (with 296k stars). However, for a small feature change it basically created 4 different plans and launched 4 agents to work in parallel (write code, write te", "context": "", "use_case": "Claude Code plugin", "switched_from": "", "story": "Spending on AI is becoming almost impossible for businesses to budget"}]}, {"text": "with some calling it overkill for small changes or no longer needed with newer models.", "cites": [{"id": 799483088, "item_id": 49967693, "url": "https://news.ycombinator.com/item?id=49967693", "by": "sajithdilshan", "date": "2026-10-05", "stance": "warn", "quote": "It ended up burning so much tokens and taking actually longer to finish and just could have been done without any agent. I immediately uninstall the plugin.", "excerpt": "It's not necessarily corporate, but mainly due to the fact that AI tools are quite new and we're just learning to use them efficiently. One example is that recently I used the superpowers claude code plugin. I thought it was good because it was popular on github (with 296k stars). However, for a small feature change it basically created 4 different plans and launched 4 agents to work in parallel (write code, write te", "context": "", "use_case": "Claude Code plugin", "switched_from": "", "story": "Spending on AI is becoming almost impossible for businesses to budget"}, {"id": 780366048, "item_id": 48772878, "url": "https://news.ycombinator.com/item?id=48772878", "by": "tmach32", "date": "2026-07-03", "stance": "compare", "quote": "IMO, the strongest part of Superpowers is /subagent-driven-development.", "excerpt": "I used Superpowers for a few weeks. I ran into a couple issues: * I wish I could turn it on selectively. Many of my requests do not require the \"verification before completion\" and TDD ceremony. For example, agents using stock Superpowers will go so far as to grep a file every time you ask to add something to them to verify that the edit really landed. * While I like speccing out/designing a project before implementa", "context": "used it for a few weeks", "use_case": "agentic coding workflow", "switched_from": "", "story": "Superpowers 6"}, {"id": 802370704, "item_id": 50148169, "url": "https://news.ycombinator.com/item?id=50148169", "by": "jemdiggity", "date": "2026-10-19", "stance": "neutral", "quote": "I found superpowers helpful maybe 4 months ago. The models don't seem to need it anymore.", "excerpt": "Like others I found superpowers helpful maybe 4 months ago. The models don't seem to need it anymore. I also don't have the same prompting issues the author has. Opus 5.5 tends to develop the spec, assumes I agree, and builds it. I make my sandwich and come back to a spec md, a diff, other evidence of completion. I use cc cli with a cmux-like platform it can control (ie open tabs showing specs, diffs itself).", "context": "", "use_case": "", "switched_from": "", "story": "Superlite: A Faster Alternatie to Superpowers"}]}], "praise": [{"label": "Structured workflow", "summary": "Commenters like that it organizes work into planning, review and implementation with scrutiny at each step.", "quotes": [{"id": 797687409, "item_id": 49855463, "url": "https://news.ycombinator.com/item?id=49855463", "by": "parasti", "date": "2026-09-26", "stance": "recommend", "quote": "superpowers applies a fairly high level of scrutiny at every step. It's really quite effective", "excerpt": "At work, we use Claude Code with superpowers which organizes most work into planning followed by review followed by implementation and superpowers applies a fairly high level of scrutiny at every step. It's really quite effective, I don't generally read the code that comes out. In that workflow I have never used plan mode. I don't know what I'd use it for. On my hobby projects I use Antigravity CLI and almost always ", "context": "", "use_case": "planning, review, implementation workflow", "switched_from": "", "story": "Plan mode is dead"}, {"id": 799794784, "item_id": 49987174, "url": "https://news.ycombinator.com/item?id=49987174", "by": "Merad", "date": "2026-10-07", "stance": "recommend", "quote": "Before I discovered superpowers I was manually telling claude to do these workflows, the plugin makes it easy.", "excerpt": "This repo I started yesterday has most of the patterns I like to use: https://github.com/mbcrawfo/SqlSource Really the single biggest piece I use today is using a tool like superpowers to ensure your task is well defined. Added bonuses are its TDD implementation skills and multiple review steps baked into the workflow. Before I discovered superpowers I was manually telling claude to do these workflows, the plugin mak", "context": "", "use_case": "defining tasks, TDD workflow", "switched_from": "", "story": "JetBrains reports revenue growth, net financial loss for 2025"}, {"id": 740226464, "item_id": 46264154, "url": "https://news.ycombinator.com/item?id=46264154", "by": "radio879", "date": "2025-12-14", "stance": "recommend", "quote": "I\u2019ve been using the Superpowers plan mode: it creates a detailed plan first, then there\u2019s an execute mode", "excerpt": "Lately, I\u2019ve been using Claude Code skills/plugins. One is called Superpowers (you should be able to find it by typing \u201cclaude code superpowers github\u201d), and there are many other skills installed for all sorts of things. Gotta give props to whoever came up with the skills idea at Anthropic. It\u2019s a smart way of handling context \u2014 instead of dumping pages of helper text into the prompt (for example, how to properly bui", "context": "", "use_case": "planning and executing long coding tasks", "switched_from": "", "story": "Ask HN: How can I get better at using AI for programming?"}]}, {"label": "Brainstorming skill", "summary": "The /brainstorming skill asks clarifying questions to fill gaps before building and gives consistent results.", "quotes": [{"id": 797710576, "item_id": 49856911, "url": "https://news.ycombinator.com/item?id=49856911", "by": "teliskr", "date": "2026-09-26", "stance": "recommend", "quote": "I've been using superpowers:brainstorming skill with claude for a while and have found the process to work well for me.", "excerpt": "I've been using superpowers:brainstorming skill with claude for a while and have found the process to work well for me. I generally spend 10-30 minutes typing my requirements and ideas for a feature and then I feed it into brainstorming skill which asks questions to fill in the gaps before starting to build things. I don't bother reading the spec it creates. I have been consistently happy with the results. However, s", "context": "", "use_case": "brainstorming feature specs", "switched_from": "", "story": "Plan mode is dead"}, {"id": 776135120, "item_id": 48508445, "url": "https://news.ycombinator.com/item?id=48508445", "by": "spgorbatiuk", "date": "2026-06-12", "stance": "recommend", "quote": "I really recommend the /brainstorming skill there.", "excerpt": "In an unlikely scenario you haven't stumbled upon the superpowers (https://claude.com/plugins/superpowers), I really recommend the /brainstorming skill there. The plugin itself leans into engineering domain, but this specific skill makes claude think better and bring up and ask thoughtful stuff before it gets to a solution part", "context": "", "use_case": "improving Claude's reasoning", "switched_from": "", "story": "Ask HN: Favorite prompts for improving LLM output?"}, {"id": 796403665, "item_id": 49775229, "url": "https://news.ycombinator.com/item?id=49775229", "by": "teliskr", "date": "2026-09-20", "stance": "recommend", "quote": "I use the superpowers:brainstorming skill. That has consistent process which has worked really well.", "excerpt": "I am getting really good results from claude. We have a 22-year old legacy system. The system is stable, but had issues as all legacy systems do. Claude has been great for modernizing the codebase, updating dependencies, auditing security, and rapidly adding new features. It has worked well with existing code style and patterns. Sometimes it is a little off-track, but overall it is pretty amazing. When implementing n", "context": "", "use_case": "planning features and refactors", "switched_from": "", "story": "If AI coding is lowering your code quality, you're not managing quality right"}]}, {"label": "Reliable results", "summary": "Long-time users say it works well across a variety of tasks and helps them work better in less time.", "quotes": [{"id": 800222496, "item_id": 50013906, "url": "https://news.ycombinator.com/item?id=50013906", "by": "runtime_terror", "date": "2026-10-08", "stance": "recommend", "quote": "If you're not using the Superpowers set of skills, give it a shot. It's been working really well for me on a variety of tasks.", "excerpt": "If you're not using the Superpowers set of skills, give it a shot. It's been working really well for me on a variety of tasks.", "context": "", "use_case": "AI agent skills", "switched_from": "", "story": "Why isn't the industry freaking out about DeepSeek 4.1 Flash?"}, {"id": 780296224, "item_id": 48768514, "url": "https://news.ycombinator.com/item?id=48768514", "by": "CharlesW", "date": "2026-07-02", "stance": "recommend", "quote": "Superpowers is one of the few skill/agent suites I use for all software development projects.", "excerpt": "As a long time user, I'd recommend checking out https://github.com/obra/superpowers#the-basic-workflow. If that reasonates with you, try using it to develop a few features or capabilities. It works very well for the way that I work (interactively and iteratively, not \"one-shot\"), and it helps me to better work in less time. Superpowers is one of the few skill/agent suites I use for all software development projects. ", "context": "long-time user", "use_case": "software development with agents", "switched_from": "", "story": "Superpowers 6"}, {"id": 768295232, "item_id": 48018452, "url": "https://news.ycombinator.com/item?id=48018452", "by": "tariky", "date": "2026-05-04", "stance": "recommend", "quote": "I use superpowers for several months now and it really does help.", "excerpt": "What is difference between superpowers and this? I use superpowers for several months now and it really does help. But still 90/10 rule applies, 10% of time it will produce stupid decision. So always check spec.", "context": "", "use_case": "AI coding workflow", "switched_from": "", "story": "Agent Skills"}]}, {"label": "Subagent orchestration", "summary": "It spins up subagents automatically, and one user calls /subagent-driven-development its strongest part.", "quotes": [{"id": 785761585, "item_id": 49110099, "url": "https://news.ycombinator.com/item?id=49110099", "by": "nvarsj", "date": "2026-07-30", "stance": "recommend", "quote": "Then I use superpowers and it spins up subagents automatically. Works really well", "excerpt": "tintinweb/pi-subagents. Then I use superpowers and it spins up subagents automatically. Works really well and you can customise the models and efforts you want to use with local AGENTS.md. I\u2019m primarily using gpt-5.6 and then opus for reviews.", "context": "", "use_case": "agent skills and subagents", "switched_from": "", "story": "Kimi K3-256k"}, {"id": 780366048, "item_id": 48772878, "url": "https://news.ycombinator.com/item?id=48772878", "by": "tmach32", "date": "2026-07-03", "stance": "compare", "quote": "IMO, the strongest part of Superpowers is /subagent-driven-development.", "excerpt": "I used Superpowers for a few weeks. I ran into a couple issues: * I wish I could turn it on selectively. Many of my requests do not require the \"verification before completion\" and TDD ceremony. For example, agents using stock Superpowers will go so far as to grep a file every time you ask to add something to them to verify that the edit really landed. * While I like speccing out/designing a project before implementa", "context": "used it for a few weeks", "use_case": "agentic coding workflow", "switched_from": "", "story": "Superpowers 6"}]}, {"label": "Customizable", "summary": "Its docs are easy to edit, and users run customized versions or set models per task.", "quotes": [{"id": 785935280, "item_id": 49120955, "url": "https://news.ycombinator.com/item?id=49120955", "by": "cthulberg", "date": "2026-07-31", "stance": "recommend", "quote": "I love it, the docs are easily editable", "excerpt": "https://github.com/obra/superpowers I use Opus/Sol with for /brainstorming, deepseek (on Pi) for /subagent-driven-development I love it, the docs are easily editable and when I'm ready Deepseek is faster/cheaper then everything on my coding plans.", "context": "", "use_case": "agent workflow with brainstorming and subagents", "switched_from": "", "story": "DeepSeek-V4-Flash Update"}, {"id": 757475712, "item_id": 47342232, "url": "https://news.ycombinator.com/item?id=47342232", "by": "notpublic", "date": "2026-03-11", "stance": "recommend", "quote": "I use it as well (a customized version suited for my worflow). It is indeed the GOAT.", "excerpt": "Thanks! I assume you are referring to this https://github.com/obra/superpowers I use it as well (a customized version suited for my worflow). It is indeed the GOAT.", "context": "", "use_case": "coding agent plugin", "switched_from": "", "story": "Show HN: Open-source browser for AI agents (~90% on Mind2Web)"}, {"id": 785761585, "item_id": 49110099, "url": "https://news.ycombinator.com/item?id=49110099", "by": "nvarsj", "date": "2026-07-30", "stance": "recommend", "quote": "Then I use superpowers and it spins up subagents automatically. Works really well", "excerpt": "tintinweb/pi-subagents. Then I use superpowers and it spins up subagents automatically. Works really well and you can customise the models and efforts you want to use with local AGENTS.md. I\u2019m primarily using gpt-5.6 and then opus for reviews.", "context": "", "use_case": "agent skills and subagents", "switched_from": "", "story": "Kimi K3-256k"}]}, {"label": "Cross-agent consistency", "summary": "Installed in both Codex and Claude Code, it produces remarkably similar behavior.", "quotes": [{"id": 767335808, "item_id": 47958488, "url": "https://news.ycombinator.com/item?id=47958488", "by": "philipbjorge", "date": "2026-04-29", "stance": "recommend", "quote": "you could install https://github.com/obra/superpowers/ into both Codex and Claude Code", "excerpt": "I haven't really shared what I use, I'm still deciding if that's something I want to do. To get an idea of what I'm talking about, you could install https://github.com/obra/superpowers/ into both Codex and Claude Code -- You'll find that the behavior is remarkably similar if you A/B compare them on the same problems. CC occasionally misses things that Codex gets and vice versa. Overall the output structure and final ", "context": "", "use_case": "harmonizing coding agent behavior", "switched_from": "", "story": "Claude.ai and API Unavailable"}]}], "caveats": [{"label": "Token usage", "summary": "Many commenters say it burns through tokens much faster.", "quotes": [{"id": 802445664, "item_id": 50152854, "url": "https://news.ycombinator.com/item?id=50152854", "by": "moondowner", "date": "2026-10-19", "stance": "recommend", "quote": "I've been using Superpowers for months, it's great, only downside is the extensive token usage.", "excerpt": "I've been using Superpowers for months, it's great, only downside is the extensive token usage. When I saw the title Superlite, I thought it's a Superpowers fork optimised for less token usage...", "context": "", "use_case": "AI coding workflow", "switched_from": "", "story": "Superlite: A Faster Alternatie to Superpowers"}, {"id": 800237824, "item_id": 50014864, "url": "https://news.ycombinator.com/item?id=50014864", "by": "tombert", "date": "2026-10-09", "stance": "warn", "quote": "I don't feel like it gave me significantly better results than just raw-dogging it.  It did however burn through my tokens significantly faster.", "excerpt": "I used Superpowers for awhile, but I don't feel like it gave me significantly better results than just raw-dogging it. It did however burn through my tokens significantly faster. Maybe I'll come to miss it now that I removed it, but I certainly don't yet.", "context": "", "use_case": "AI coding skills", "switched_from": "", "story": "Why isn't the industry freaking out about DeepSeek 4.1 Flash?"}, {"id": 780294288, "item_id": 48768393, "url": "https://news.ycombinator.com/item?id=48768393", "by": "smusamashah", "date": "2026-07-02", "stance": "neutral", "quote": "I only found it burning too many tokens to do too little. I guess Superpowers is useful only in hands that know how to manipulate it.", "excerpt": "Where I $work, someone used Superpowers to pull off two big projects that before AI have always been left untouched because of the effort and time required. One was about unifying lots of duplicating (but kot exactly) libraries, and another to convert our bespoke shell scripts used throughout deployment pipeline to ansible. When I used it though , I only found it burning too many tokens to do too little. I guess Supe", "context": "", "use_case": "AI-assisted dev projects", "switched_from": "", "story": "Superpowers 6"}]}, {"label": "Mixed results", "summary": "Some found results no better or even worse than without it, especially with some models.", "quotes": [{"id": 800237824, "item_id": 50014864, "url": "https://news.ycombinator.com/item?id=50014864", "by": "tombert", "date": "2026-10-09", "stance": "warn", "quote": "I don't feel like it gave me significantly better results than just raw-dogging it.  It did however burn through my tokens significantly faster.", "excerpt": "I used Superpowers for awhile, but I don't feel like it gave me significantly better results than just raw-dogging it. It did however burn through my tokens significantly faster. Maybe I'll come to miss it now that I removed it, but I certainly don't yet.", "context": "", "use_case": "AI coding skills", "switched_from": "", "story": "Why isn't the industry freaking out about DeepSeek 4.1 Flash?"}, {"id": 762000240, "item_id": 47625015, "url": "https://news.ycombinator.com/item?id=47625015", "by": "d--b", "date": "2026-04-03", "stance": "warn", "quote": "I personally don't like superpowers very much.", "excerpt": "I personally don't like superpowers very much. My boss does. I think Claude makes more mistakes when using superpowers than when not. but maybe it's my fault. I recommend trying. it doesn't hurt. Just don't believe it's a silver bullet. It's still the same Claude. The way I use Claude is quite similar to what super powers does under the hood anyways. Like I always ask things like this: \"if I want to do X, what questi", "context": "", "use_case": "Claude Code plugin", "switched_from": "", "story": "A Rave Review of Superpowers (For Claude Code)"}, {"id": 780312848, "item_id": 48769553, "url": "https://news.ycombinator.com/item?id=48769553", "by": "jorl17", "date": "2026-07-02", "stance": "neutral", "quote": "If I use it with other models it's terrible. With composer2.5 it is slightly better, though not much.", "excerpt": "I only use superpowers when I want to stick with composer2.5(fast) for everything. If I use it with other models it's terrible. With composer2.5 it is slightly better, though not much.", "context": "", "use_case": "AI coding workflow", "switched_from": "", "story": "Superpowers 6"}]}, {"label": "Too much ceremony", "summary": "Its multi-agent, verification and TDD steps can be overkill and slow for small changes.", "quotes": [{"id": 799483088, "item_id": 49967693, "url": "https://news.ycombinator.com/item?id=49967693", "by": "sajithdilshan", "date": "2026-10-05", "stance": "warn", "quote": "It ended up burning so much tokens and taking actually longer to finish and just could have been done without any agent. I immediately uninstall the plugin.", "excerpt": "It's not necessarily corporate, but mainly due to the fact that AI tools are quite new and we're just learning to use them efficiently. One example is that recently I used the superpowers claude code plugin. I thought it was good because it was popular on github (with 296k stars). However, for a small feature change it basically created 4 different plans and launched 4 agents to work in parallel (write code, write te", "context": "", "use_case": "Claude Code plugin", "switched_from": "", "story": "Spending on AI is becoming almost impossible for businesses to budget"}, {"id": 780366048, "item_id": 48772878, "url": "https://news.ycombinator.com/item?id=48772878", "by": "tmach32", "date": "2026-07-03", "stance": "compare", "quote": "IMO, the strongest part of Superpowers is /subagent-driven-development.", "excerpt": "I used Superpowers for a few weeks. I ran into a couple issues: * I wish I could turn it on selectively. Many of my requests do not require the \"verification before completion\" and TDD ceremony. For example, agents using stock Superpowers will go so far as to grep a file every time you ask to add something to them to verify that the edit really landed. * While I like speccing out/designing a project before implementa", "context": "used it for a few weeks", "use_case": "agentic coding workflow", "switched_from": "", "story": "Superpowers 6"}, {"id": 753716578, "item_id": 47107286, "url": "https://news.ycombinator.com/item?id=47107286", "by": "brandall10", "date": "2026-02-21", "stance": "warn", "quote": "I gave superpowers a shot... which was fine, but still too much.", "excerpt": "I go a bit further than this and have had incredible success with 3 doc types: - Specs: these are generally static, but updatable as the project evolves. And they're broken out to an index file that gives a project overview, a high-level arch file, and files for all the main modules. Roughly ~1k lines of spec for 10k lines of code - Plans: these are the output of a planning session with an LLM. They point to the asso", "context": "", "use_case": "spec generation", "switched_from": "", "story": "How I use Claude Code: Separation of planning and execution"}]}, {"label": "Shallow TDD", "summary": "Its enforced TDD can fill test suites with low-value tests, and some don't want TDD at all.", "quotes": [{"id": 793710736, "item_id": 49606921, "url": "https://news.ycombinator.com/item?id=49606921", "by": "ivanzhaowy123", "date": "2026-09-08", "stance": "neutral", "quote": "when I use the Superpowers skill set, the agent proactively adopts TDD for every new feature.", "excerpt": "In my experience, agents often think of more edge cases than humans when writing unit tests. But under the guidance of certain skills, they can become mechanical and lose sight of the business logic. For example, when I use the Superpowers skill set, the agent proactively adopts TDD for every new feature. But its understanding of testing often stays superficial: if the user asks for a screen with a \u201cSend\u201d button, it ", "context": "", "use_case": "agent TDD workflow", "switched_from": "", "story": "How well do agents use test/verification techniques?"}, {"id": 780302352, "item_id": 48768897, "url": "https://news.ycombinator.com/item?id=48768897", "by": "mempko", "date": "2026-07-02", "stance": "compare", "quote": "This is great in concept but what prevents me from using it is TDD.", "excerpt": "This is great in concept but what prevents me from using it is TDD. I don't want to waste tokens on producing code that doesn't ship to the end user. Design by Contract is a far superior approach. If you've never heard of Design by Contract I don't blame you, our culture really failed to bring it mainstream. But I swear by it and it gives me real superpowers. Maybe I should fork this and gut the TDD part and replace ", "context": "", "use_case": "agentic coding workflow", "switched_from": "", "story": "Superpowers 6"}]}, {"label": "Less needed now", "summary": "Some say newer models no longer need it and the skills just add clutter.", "quotes": [{"id": 802370704, "item_id": 50148169, "url": "https://news.ycombinator.com/item?id=50148169", "by": "jemdiggity", "date": "2026-10-19", "stance": "neutral", "quote": "I found superpowers helpful maybe 4 months ago. The models don't seem to need it anymore.", "excerpt": "Like others I found superpowers helpful maybe 4 months ago. The models don't seem to need it anymore. I also don't have the same prompting issues the author has. Opus 5.5 tends to develop the spec, assumes I agree, and builds it. I make my sandwich and come back to a spec md, a diff, other evidence of completion. I use cc cli with a cmux-like platform it can control (ie open tabs showing specs, diffs itself).", "context": "", "use_case": "", "switched_from": "", "story": "Superlite: A Faster Alternatie to Superpowers"}, {"id": 799361936, "item_id": 49960121, "url": "https://news.ycombinator.com/item?id=49960121", "by": "kaydub", "date": "2026-10-05", "stance": "warn", "quote": "They're garbage. That shit was maybe helpful a year ago. It's now just cluttering up space.", "excerpt": "Okay, so using superpowers. Have you actually looked at the superpowers? They're garbage. That shit was maybe helpful a year ago. It's now just cluttering up space. I think a lot of people are stuck on patterns that \"used to work\" (and I use that loosely because I really feel like LLMs are getting work done IN SPITE of using all these additional skills/plugins/etc). Go ahead, delete all that shit and try building wit", "context": "", "use_case": "agent skills", "switched_from": "", "story": "Agents don't need memory, they need documentation"}]}, {"label": "Needs review", "summary": "Users still check its specs because it sometimes makes poor decisions.", "quotes": [{"id": 768295232, "item_id": 48018452, "url": "https://news.ycombinator.com/item?id=48018452", "by": "tariky", "date": "2026-05-04", "stance": "recommend", "quote": "I use superpowers for several months now and it really does help.", "excerpt": "What is difference between superpowers and this? I use superpowers for several months now and it really does help. But still 90/10 rule applies, 10% of time it will produce stupid decision. So always check spec.", "context": "", "use_case": "AI coding workflow", "switched_from": "", "story": "Agent Skills"}]}], "who": [{"by": "cg-enterprise", "context": "built own custom workflow on top of SDD frameworks", "url": "https://news.ycombinator.com/item?id=49737218", "date": "2026-09-17"}, {"by": "ksaun", "context": "building a game called Vestiges", "url": "https://news.ycombinator.com/item?id=48902377", "date": "2026-07-13"}, {"by": "CharlesW", "context": "long-time user", "url": "https://news.ycombinator.com/item?id=48768514", "date": "2026-07-02"}], "tags": ["for Claude Code users", "for spec-driven development", "for large features and refactors", "for iterative, interactive work", "works with Codex"], "use_cases": ["ai coding workflow", "claude code plugin", "agent skills", "ai coding skills", "ai agent skills", "claude code skills"], "compared": [{"slug": "claude-code", "name": "Claude Code", "co": 10, "switched_from": 0, "switched_to": 0}, {"slug": "claude-opus", "name": "Claude Opus", "co": 7, "switched_from": 0, "switched_to": 0}, {"slug": "claude", "name": "Claude", "co": 6, "switched_from": 0, "switched_to": 0}, {"slug": "codex", "name": "Codex", "co": 4, "switched_from": 0, "switched_to": 0}, {"slug": "openspec", "name": "OpenSpec", "co": 4, "switched_from": 0, "switched_to": 0}, {"slug": "ponytail", "name": "ponytail", "co": 3, "switched_from": 0, "switched_to": 0}, {"slug": "gsd", "name": "GSD", "co": 3, "switched_from": 0, "switched_to": 0}, {"slug": "spec-kit", "name": "spec-kit", "co": 3, "switched_from": 0, "switched_to": 0}], "makers": [], "threads": [{"url": "https://news.ycombinator.com/item?id=48739459", "title": "Superpowers 6", "mentions": 7}, {"url": "https://news.ycombinator.com/item?id=50147492", "title": "Superlite: A Faster Alternatie to Superpowers", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=50000488", "title": "Why isn't the industry freaking out about DeepSeek 4.1 Flash?", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=49945933", "title": "Agents don't need memory, they need documentation", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=49840054", "title": "Plan mode is dead", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=49977072", "title": "JetBrains reports revenue growth, net financial loss for 2025", "mentions": 1}], "sentiment_by_month": {"2025-11": {"neutral": 1}, "2025-12": {"recommend": 1}, "2026-01": {"recommend": 1}, "2026-02": {"recommend": 1, "warn": 1}, "2026-03": {"recommend": 2}, "2026-04": {"warn": 2, "recommend": 2}, "2026-05": {"recommend": 1}, "2026-06": {"recommend": 2, "neutral": 1}, "2026-07": {"neutral": 3, "recommend": 4, "compare": 3, "warn": 1}, "2026-08": {"neutral": 4, "recommend": 1, "warn": 1}, "2026-09": {"neutral": 2, "recommend": 4}, "2026-10": {"neutral": 2, "warn": 3, "recommend": 3}}, "quotes": [{"id": 802445664, "item_id": 50152854, "url": "https://news.ycombinator.com/item?id=50152854", "by": "moondowner", "date": "2026-10-19", "stance": "recommend", "quote": "I've been using Superpowers for months, it's great, only downside is the extensive token usage.", "excerpt": "I've been using Superpowers for months, it's great, only downside is the extensive token usage. When I saw the title Superlite, I thought it's a Superpowers fork optimised for less token usage...", "context": "", "use_case": "AI coding workflow", "switched_from": "", "story": "Superlite: A Faster Alternatie to Superpowers"}, {"id": 802370704, "item_id": 50148169, "url": "https://news.ycombinator.com/item?id=50148169", "by": "jemdiggity", "date": "2026-10-19", "stance": "neutral", "quote": "I found superpowers helpful maybe 4 months ago. The models don't seem to need it anymore.", "excerpt": "Like others I found superpowers helpful maybe 4 months ago. The models don't seem to need it anymore. I also don't have the same prompting issues the author has. Opus 5.5 tends to develop the spec, assumes I agree, and builds it. I make my sandwich and come back to a spec md, a diff, other evidence of completion. I use cc cli with a cmux-like platform it can control (ie open tabs showing specs, diffs itself).", "context": "", "use_case": "", "switched_from": "", "story": "Superlite: A Faster Alternatie to Superpowers"}, {"id": 800237824, "item_id": 50014864, "url": "https://news.ycombinator.com/item?id=50014864", "by": "tombert", "date": "2026-10-09", "stance": "warn", "quote": "I don't feel like it gave me significantly better results than just raw-dogging it.  It did however burn through my tokens significantly faster.", "excerpt": "I used Superpowers for awhile, but I don't feel like it gave me significantly better results than just raw-dogging it. It did however burn through my tokens significantly faster. Maybe I'll come to miss it now that I removed it, but I certainly don't yet.", "context": "", "use_case": "AI coding skills", "switched_from": "", "story": "Why isn't the industry freaking out about DeepSeek 4.1 Flash?"}, {"id": 800222496, "item_id": 50013906, "url": "https://news.ycombinator.com/item?id=50013906", "by": "runtime_terror", "date": "2026-10-08", "stance": "recommend", "quote": "If you're not using the Superpowers set of skills, give it a shot. It's been working really well for me on a variety of tasks.", "excerpt": "If you're not using the Superpowers set of skills, give it a shot. It's been working really well for me on a variety of tasks.", "context": "", "use_case": "AI agent skills", "switched_from": "", "story": "Why isn't the industry freaking out about DeepSeek 4.1 Flash?"}, {"id": 799794784, "item_id": 49987174, "url": "https://news.ycombinator.com/item?id=49987174", "by": "Merad", "date": "2026-10-07", "stance": "recommend", "quote": "Before I discovered superpowers I was manually telling claude to do these workflows, the plugin makes it easy.", "excerpt": "This repo I started yesterday has most of the patterns I like to use: https://github.com/mbcrawfo/SqlSource Really the single biggest piece I use today is using a tool like superpowers to ensure your task is well defined. Added bonuses are its TDD implementation skills and multiple review steps baked into the workflow. Before I discovered superpowers I was manually telling claude to do these workflows, the plugin mak", "context": "", "use_case": "defining tasks, TDD workflow", "switched_from": "", "story": "JetBrains reports revenue growth, net financial loss for 2025"}, {"id": 799483088, "item_id": 49967693, "url": "https://news.ycombinator.com/item?id=49967693", "by": "sajithdilshan", "date": "2026-10-05", "stance": "warn", "quote": "It ended up burning so much tokens and taking actually longer to finish and just could have been done without any agent. I immediately uninstall the plugin.", "excerpt": "It's not necessarily corporate, but mainly due to the fact that AI tools are quite new and we're just learning to use them efficiently. One example is that recently I used the superpowers claude code plugin. I thought it was good because it was popular on github (with 296k stars). However, for a small feature change it basically created 4 different plans and launched 4 agents to work in parallel (write code, write te", "context": "", "use_case": "Claude Code plugin", "switched_from": "", "story": "Spending on AI is becoming almost impossible for businesses to budget"}, {"id": 799361936, "item_id": 49960121, "url": "https://news.ycombinator.com/item?id=49960121", "by": "kaydub", "date": "2026-10-05", "stance": "warn", "quote": "They're garbage. That shit was maybe helpful a year ago. It's now just cluttering up space.", "excerpt": "Okay, so using superpowers. Have you actually looked at the superpowers? They're garbage. That shit was maybe helpful a year ago. It's now just cluttering up space. I think a lot of people are stuck on patterns that \"used to work\" (and I use that loosely because I really feel like LLMs are getting work done IN SPITE of using all these additional skills/plugins/etc). Go ahead, delete all that shit and try building wit", "context": "", "use_case": "agent skills", "switched_from": "", "story": "Agents don't need memory, they need documentation"}, {"id": 799358928, "item_id": 49959933, "url": "https://news.ycombinator.com/item?id=49959933", "by": "marcus_holmes", "date": "2026-10-05", "stance": "neutral", "quote": "if I'm using superpowers, it generates two, the design doc and the implementation plan", "excerpt": "I have a \"docs\" folder in every project. Every iteration on the project generates a plan (if I'm using superpowers, it generates two, the design doc and the implementation plan). Those go into docs. Inside the docs folder is an archive folder. Stale and completed plans and docs get moved there. I generate a schema.md that describes the database schema in LLM-friendly terms that lives in docs. If there's a gnarly desi", "context": "", "use_case": "planning for coding agents", "switched_from": "", "story": "Agents don't need memory, they need documentation"}, {"id": 797710576, "item_id": 49856911, "url": "https://news.ycombinator.com/item?id=49856911", "by": "teliskr", "date": "2026-09-26", "stance": "recommend", "quote": "I've been using superpowers:brainstorming skill with claude for a while and have found the process to work well for me.", "excerpt": "I've been using superpowers:brainstorming skill with claude for a while and have found the process to work well for me. I generally spend 10-30 minutes typing my requirements and ideas for a feature and then I feed it into brainstorming skill which asks questions to fill in the gaps before starting to build things. I don't bother reading the spec it creates. I have been consistently happy with the results. However, s", "context": "", "use_case": "brainstorming feature specs", "switched_from": "", "story": "Plan mode is dead"}, {"id": 797687409, "item_id": 49855463, "url": "https://news.ycombinator.com/item?id=49855463", "by": "parasti", "date": "2026-09-26", "stance": "recommend", "quote": "superpowers applies a fairly high level of scrutiny at every step. It's really quite effective", "excerpt": "At work, we use Claude Code with superpowers which organizes most work into planning followed by review followed by implementation and superpowers applies a fairly high level of scrutiny at every step. It's really quite effective, I don't generally read the code that comes out. In that workflow I have never used plan mode. I don't know what I'd use it for. On my hobby projects I use Antigravity CLI and almost always ", "context": "", "use_case": "planning, review, implementation workflow", "switched_from": "", "story": "Plan mode is dead"}, {"id": 796403665, "item_id": 49775229, "url": "https://news.ycombinator.com/item?id=49775229", "by": "teliskr", "date": "2026-09-20", "stance": "recommend", "quote": "I use the superpowers:brainstorming skill. That has consistent process which has worked really well.", "excerpt": "I am getting really good results from claude. We have a 22-year old legacy system. The system is stable, but had issues as all legacy systems do. Claude has been great for modernizing the codebase, updating dependencies, auditing security, and rapidly adding new features. It has worked well with existing code style and patterns. Sometimes it is a little off-track, but overall it is pretty amazing. When implementing n", "context": "", "use_case": "planning features and refactors", "switched_from": "", "story": "If AI coding is lowering your code quality, you're not managing quality right"}, {"id": 796397252, "item_id": 49774828, "url": "https://news.ycombinator.com/item?id=49774828", "by": "JoyenBenitto", "date": "2026-09-20", "stance": "recommend", "quote": "plugins like GSD and superpower which I am accustomed too", "excerpt": "Yeah I have been using it to automate almost everything, even my gitlab ci's sometimes when those ci fails I just ask claude to use glab cli to figure things out on it's own. I have used codex and antigravity but not a huge fan of them. I feel my I am too much in favour of claude also because of plugins like GSD and superpower which I am accustomed too.", "context": "", "use_case": "Claude Code plugin", "switched_from": "", "story": "Ask HN: What happens when Claude Code goes away?"}, {"id": 795795488, "item_id": 49737218, "url": "https://news.ycombinator.com/item?id=49737218", "by": "cg-enterprise", "date": "2026-09-17", "stance": "recommend", "quote": "for SDD I prefer to use Superpowers and/or Matt Pocock's skills", "excerpt": "It's interesting to see the divergence of opinions on these frameworks and also at least anecdotally how many people roll their own custom implementations of workflow management on top of their favourite SDD framework. I also ended up doing my own thing mainly to address several omissions in the existing frameworks (for SDD I prefer to use Superpowers and/or Matt Pocock's skills): 1. Artifact staleness and tracking -", "context": "built own custom workflow on top of SDD frameworks", "use_case": "spec-driven development", "switched_from": "", "story": "OpenSpec \u2013 A lightweight and configurable AI spec framework"}, {"id": 793710736, "item_id": 49606921, "url": "https://news.ycombinator.com/item?id=49606921", "by": "ivanzhaowy123", "date": "2026-09-08", "stance": "neutral", "quote": "when I use the Superpowers skill set, the agent proactively adopts TDD for every new feature.", "excerpt": "In my experience, agents often think of more edge cases than humans when writing unit tests. But under the guidance of certain skills, they can become mechanical and lose sight of the business logic. For example, when I use the Superpowers skill set, the agent proactively adopts TDD for every new feature. But its understanding of testing often stays superficial: if the user asks for a screen with a \u201cSend\u201d button, it ", "context": "", "use_case": "agent TDD workflow", "switched_from": "", "story": "How well do agents use test/verification techniques?"}, {"id": 792949586, "item_id": 49559349, "url": "https://news.ycombinator.com/item?id=49559349", "by": "kimseungyong", "date": "2026-09-04", "stance": "neutral", "quote": "I usually develop with superpowers planing.", "excerpt": "Yeah I agree. I used Fable 5.1 again and my token 30% suddenly gone. I usually develop with superpowers planing. And 30% is disappeared with only planning. I turned back to Opus directly", "context": "", "use_case": "planning", "switched_from": "", "story": "Claude Fable 5.1 and Claude Mythos 5.1"}, {"id": 792127188, "item_id": 49507949, "url": "https://news.ycombinator.com/item?id=49507949", "by": "mstaoru", "date": "2026-08-31", "stance": "neutral", "quote": "I experiment with different plan-build-evaluate frameworks like GSD, OpenSpec, Superpowers.", "excerpt": "It is very difficult. I consider myself reasonably skilled in AI usage (multiple harnesses, skills, MCP, etc). I use only the latest models: Opus 5, GPT 5.6 Sol. I experiment with different plan-build-evaluate frameworks like GSD, OpenSpec, Superpowers. I have a reasonable AGENTS.md without much cruft: test coverage ~80%, ASD-STE100 English, use uv - things like that. I can churn out single-use scripts and micro-proj", "context": "", "use_case": "plan-build-evaluate workflows", "switched_from": "", "story": "Why do I lose my passion and want to do nothing?"}, {"id": 790330801, "item_id": 49395675, "url": "https://news.ycombinator.com/item?id=49395675", "by": "sejje", "date": "2026-08-22", "stance": "neutral", "quote": "I use superpowers and ponytail a lot.", "excerpt": "It may be ipython making it work well with ds flash, too. I haven't really run many separate experiments, to be honest. I also like prime-agent's way of handling sessions better than any other harness i've used. You can run multiple agents from one instance, although the scoping could be better. But they can interact with past sessions, so preserving context isn't as important all the time. I just tell them to search", "context": "", "use_case": "agent skills", "switched_from": "", "story": "A week of using Codex more than Claude"}, {"id": 790115634, "item_id": 49382227, "url": "https://news.ycombinator.com/item?id=49382227", "by": "nikolay", "date": "2026-08-20", "stance": "warn", "quote": "With the Superpowers skill (which I am trying to get rid of)", "excerpt": "I use Claude with the Codex plugin as a critic (sometimes I switch and manually review with Claude). With the Superpowers skill (which I am trying to get rid of), it designs expensive models and implementations, uses cheaper ones for other, less important stuff, including Haiku, sometimes upgrading if it hits a roadblock, and then, finally, reviews with expensive ones. Why would I need Composer unless I'm a poor hobb", "context": "", "use_case": "agent workflow", "switched_from": "", "story": "Git at Any Scale"}, {"id": 789436496, "item_id": 49339781, "url": "https://news.ycombinator.com/item?id=49339781", "by": "qup", "date": "2026-08-17", "stance": "recommend", "quote": "I use the superpowers plug-in which I think guides the planning.", "excerpt": "I've found it to be great for planning code changes (or new projects). I use the superpowers plug-in which I think guides the planning. Then I switch models (to luna) before implementation. I find this combo nearly always does what I want. I also use a skill called ponytail, its goal is to keep things terse and edits small. It may have contributed to the successes above. I like that skills are easy to try out, too.", "context": "", "use_case": "planning code changes", "switched_from": "", "story": "GPT-5.6 Sol Pricing Cut by 50%"}, {"id": 787612465, "item_id": 49225779, "url": "https://news.ycombinator.com/item?id=49225779", "by": "Ovid", "date": "2026-08-08", "stance": "neutral", "quote": "my personal setup is Claude Code with Superpowers, Ponytail, PAAD", "excerpt": "Many people trying to develop software treat AI as magic pixie dust they can sprinkle on a problem to make it go away. For coding, this as \"AI-led engineering.\" The literature is rife with the failures of this. Instead, I argue that we need to bring back well-known engineering best practices and treat AI as a tool to support them. This is \"engineering-led AI.\" I created a free, online tutorial to explain this approac", "context": "", "use_case": "Claude Code skills", "switched_from": "", "story": "How to write production-quality code with AI"}, {"id": 786495124, "item_id": 49155945, "url": "https://news.ycombinator.com/item?id=49155945", "by": "nerdalytics", "date": "2026-08-03", "stance": "neutral", "quote": "Superpowers works, because of the same Must use wording.", "excerpt": "I used Qwen3.8 Max Preview for 2 weeks and can't tell if I truly miss Fable. Qwen doesn't overuse em-dashes, weird sentence structures with bold text: normal text. In my private GitHub Repo Qwen finished as much tasks as Fable did. Without hitting 5-hour session limits. Qwen's token-per-second performance fluctuated greatly. From 20 tps up to 80 tps. Qwen failed on some \"reading between the lines\", but so did Fable. ", "context": "", "use_case": "agent skills", "switched_from": "", "story": "Qwen3.8-Max: A New Bar for Coding and Cowork"}, {"id": 785935280, "item_id": 49120955, "url": "https://news.ycombinator.com/item?id=49120955", "by": "cthulberg", "date": "2026-07-31", "stance": "recommend", "quote": "I love it, the docs are easily editable", "excerpt": "https://github.com/obra/superpowers I use Opus/Sol with for /brainstorming, deepseek (on Pi) for /subagent-driven-development I love it, the docs are easily editable and when I'm ready Deepseek is faster/cheaper then everything on my coding plans.", "context": "", "use_case": "agent workflow with brainstorming and subagents", "switched_from": "", "story": "DeepSeek-V4-Flash Update"}, {"id": 785898592, "item_id": 49118662, "url": "https://news.ycombinator.com/item?id=49118662", "by": "glitchc", "date": "2026-07-30", "stance": "neutral", "quote": "", "excerpt": "I use my local GPU. Yes, I realize it's not exactly a superpower, but it's technology that's pretty darn close to one.", "context": "", "use_case": "agent skills setup", "switched_from": "", "story": "Does Speaking to Agents Like Cavemen Save 65% of Tokens? We Test"}, {"id": 785761585, "item_id": 49110099, "url": "https://news.ycombinator.com/item?id=49110099", "by": "nvarsj", "date": "2026-07-30", "stance": "recommend", "quote": "Then I use superpowers and it spins up subagents automatically. Works really well", "excerpt": "tintinweb/pi-subagents. Then I use superpowers and it spins up subagents automatically. Works really well and you can customise the models and efforts you want to use with local AGENTS.md. I\u2019m primarily using gpt-5.6 and then opus for reviews.", "context": "", "use_case": "agent skills and subagents", "switched_from": "", "story": "Kimi K3-256k"}, {"id": 784408433, "item_id": 49025527, "url": "https://news.ycombinator.com/item?id=49025527", "by": "prabhanjana_c", "date": "2026-07-23", "stance": "compare", "quote": "using some skills from https://github.com/obra/superpowers", "excerpt": "Interesting. Still wondering how this is different from a structured app building with let us say https://github.com/github/spec-kit, or using some skills from https://github.com/obra/superpowers. Will try it out.", "context": "", "use_case": "AI coding skills", "switched_from": "", "story": "Show HN: A primer for non-coders to design software via structured AI dialogue"}, {"id": 782438033, "item_id": 48902377, "url": "https://news.ycombinator.com/item?id=48902377", "by": "ksaun", "date": "2026-07-13", "stance": "recommend", "quote": "For any significant new feature or work, I use /brainstorming (superpowers plug-in).", "excerpt": "Overall, I've been very pleased, though I have seen inconsistency like you mention. First, some background: At the start, I used Sonnet, but after a couple weeks, I'd switched to mostly Opus. I've used Fable when available. For any significant new feature or work, I use /brainstorming (superpowers plug-in). For Vestiges, I haven't written any design documentation per se. I'll start off a /brainstorming with a hefty p", "context": "building a game called Vestiges", "use_case": "feature brainstorming and design", "switched_from": "", "story": "Ask HN: What Are You Working On? (July 2026)"}, {"id": 780366048, "item_id": 48772878, "url": "https://news.ycombinator.com/item?id=48772878", "by": "tmach32", "date": "2026-07-03", "stance": "compare", "quote": "IMO, the strongest part of Superpowers is /subagent-driven-development.", "excerpt": "I used Superpowers for a few weeks. I ran into a couple issues: * I wish I could turn it on selectively. Many of my requests do not require the \"verification before completion\" and TDD ceremony. For example, agents using stock Superpowers will go so far as to grep a file every time you ask to add something to them to verify that the edit really landed. * While I like speccing out/designing a project before implementa", "context": "used it for a few weeks", "use_case": "agentic coding workflow", "switched_from": "", "story": "Superpowers 6"}, {"id": 780316017, "item_id": 48769751, "url": "https://news.ycombinator.com/item?id=48769751", "by": "arikrahman", "date": "2026-07-02", "stance": "warn", "quote": "Likewise.", "excerpt": "Likewise. I use Reasonix with planning-with-files and that's been better than even openspec. Cache hits with the help of Reasonix on Deepseek make requests practically free. And that's unsubsidized, with American providers like Digital Ocean or cloudflare.", "context": "", "use_case": "AI coding workflow", "switched_from": "", "story": "Superpowers 6"}, {"id": 780312848, "item_id": 48769553, "url": "https://news.ycombinator.com/item?id=48769553", "by": "jorl17", "date": "2026-07-02", "stance": "neutral", "quote": "If I use it with other models it's terrible. With composer2.5 it is slightly better, though not much.", "excerpt": "I only use superpowers when I want to stick with composer2.5(fast) for everything. If I use it with other models it's terrible. With composer2.5 it is slightly better, though not much.", "context": "", "use_case": "AI coding workflow", "switched_from": "", "story": "Superpowers 6"}, {"id": 780302352, "item_id": 48768897, "url": "https://news.ycombinator.com/item?id=48768897", "by": "mempko", "date": "2026-07-02", "stance": "compare", "quote": "This is great in concept but what prevents me from using it is TDD.", "excerpt": "This is great in concept but what prevents me from using it is TDD. I don't want to waste tokens on producing code that doesn't ship to the end user. Design by Contract is a far superior approach. If you've never heard of Design by Contract I don't blame you, our culture really failed to bring it mainstream. But I swear by it and it gives me real superpowers. Maybe I should fork this and gut the TDD part and replace ", "context": "", "use_case": "agentic coding workflow", "switched_from": "", "story": "Superpowers 6"}, {"id": 780296224, "item_id": 48768514, "url": "https://news.ycombinator.com/item?id=48768514", "by": "CharlesW", "date": "2026-07-02", "stance": "recommend", "quote": "Superpowers is one of the few skill/agent suites I use for all software development projects.", "excerpt": "As a long time user, I'd recommend checking out https://github.com/obra/superpowers#the-basic-workflow. If that reasonates with you, try using it to develop a few features or capabilities. It works very well for the way that I work (interactively and iteratively, not \"one-shot\"), and it helps me to better work in less time. Superpowers is one of the few skill/agent suites I use for all software development projects. ", "context": "long-time user", "use_case": "software development with agents", "switched_from": "", "story": "Superpowers 6"}, {"id": 780294288, "item_id": 48768393, "url": "https://news.ycombinator.com/item?id=48768393", "by": "smusamashah", "date": "2026-07-02", "stance": "neutral", "quote": "I only found it burning too many tokens to do too little. I guess Superpowers is useful only in hands that know how to manipulate it.", "excerpt": "Where I $work, someone used Superpowers to pull off two big projects that before AI have always been left untouched because of the effort and time required. One was about unifying lots of duplicating (but kot exactly) libraries, and another to convert our bespoke shell scripts used throughout deployment pipeline to ansible. When I used it though , I only found it burning too many tokens to do too little. I guess Supe", "context": "", "use_case": "AI-assisted dev projects", "switched_from": "", "story": "Superpowers 6"}, {"id": 779840096, "item_id": 48740006, "url": "https://news.ycombinator.com/item?id=48740006", "by": "dmix", "date": "2026-06-30", "stance": "neutral", "quote": "Seems to be an opinionated collection of skills for dev work", "excerpt": "Neither the article or the corporate blog post explains what Superpowers is. Seems to be an opinionated collection of skills for dev work https://github.com/obra/superpowers", "context": "", "use_case": "", "switched_from": "", "story": "Superpowers 6"}, {"id": 776973920, "item_id": 48560870, "url": "https://news.ycombinator.com/item?id=48560870", "by": "amai", "date": "2026-06-16", "stance": "recommend", "quote": "See also https://github.com/obra/superpowers", "excerpt": "See also https://github.com/obra/superpowers and https://github.com/Raxyl00/titan-agent-cli", "context": "", "use_case": "AI agent skills", "switched_from": "", "story": "Ponytail \u2013 make your AI agent think like the laziest senior dev in the room"}, {"id": 776135120, "item_id": 48508445, "url": "https://news.ycombinator.com/item?id=48508445", "by": "spgorbatiuk", "date": "2026-06-12", "stance": "recommend", "quote": "I really recommend the /brainstorming skill there.", "excerpt": "In an unlikely scenario you haven't stumbled upon the superpowers (https://claude.com/plugins/superpowers), I really recommend the /brainstorming skill there. The plugin itself leans into engineering domain, but this specific skill makes claude think better and bring up and ask thoughtful stuff before it gets to a solution part", "context": "", "use_case": "improving Claude's reasoning", "switched_from": "", "story": "Ask HN: Favorite prompts for improving LLM output?"}, {"id": 768295232, "item_id": 48018452, "url": "https://news.ycombinator.com/item?id=48018452", "by": "tariky", "date": "2026-05-04", "stance": "recommend", "quote": "I use superpowers for several months now and it really does help.", "excerpt": "What is difference between superpowers and this? I use superpowers for several months now and it really does help. But still 90/10 rule applies, 10% of time it will produce stupid decision. So always check spec.", "context": "", "use_case": "AI coding workflow", "switched_from": "", "story": "Agent Skills"}, {"id": 767335808, "item_id": 47958488, "url": "https://news.ycombinator.com/item?id=47958488", "by": "philipbjorge", "date": "2026-04-29", "stance": "recommend", "quote": "you could install https://github.com/obra/superpowers/ into both Codex and Claude Code", "excerpt": "I haven't really shared what I use, I'm still deciding if that's something I want to do. To get an idea of what I'm talking about, you could install https://github.com/obra/superpowers/ into both Codex and Claude Code -- You'll find that the behavior is remarkably similar if you A/B compare them on the same problems. CC occasionally misses things that Codex gets and vice versa. Overall the output structure and final ", "context": "", "use_case": "harmonizing coding agent behavior", "switched_from": "", "story": "Claude.ai and API Unavailable"}, {"id": 767051905, "item_id": 47940744, "url": "https://news.ycombinator.com/item?id=47940744", "by": "bushido", "date": "2026-04-28", "stance": "warn", "quote": "Oh, I use GSD or I use Superpowers and my results have gotten worse", "excerpt": "I think the main thing which a lot of these articles miss is it's not just your Agents.md which can give you a model upgrade or the inverse. But everything your harness looks at could be this. So the skills in your code base, the commands that you've added, the memories that were auto created, they all work towards improving or completely destroying your productivity. And most of it is hidden. You hear people talk ab", "context": "", "use_case": "AI coding agent workflow", "switched_from": "", "story": "A good AGENTS.md is a model upgrade. A bad one is worse than no docs at all"}, {"id": 762856688, "item_id": 47678543, "url": "https://news.ycombinator.com/item?id=47678543", "by": "julian37", "date": "2026-04-07", "stance": "recommend", "quote": "I recommend brainstorming everything with https://github.com/obra/superpowers", "excerpt": "Use planning+execution rather than one-shotting, it'll let you push back on stuff like this. I recommend brainstorming everything with https://github.com/obra/superpowers, at least to start with. Then work on making sure the LLM has all the info it needs. In this example it sounds like perhaps your hypothetical data model would need to be better typed and/or documented. But yeah as of today it won't pick up on smells", "context": "", "use_case": "planning with Claude Code", "switched_from": "", "story": "Claude Code is locking people out for hours"}, {"id": 762000240, "item_id": 47625015, "url": "https://news.ycombinator.com/item?id=47625015", "by": "d--b", "date": "2026-04-03", "stance": "warn", "quote": "I personally don't like superpowers very much.", "excerpt": "I personally don't like superpowers very much. My boss does. I think Claude makes more mistakes when using superpowers than when not. but maybe it's my fault. I recommend trying. it doesn't hurt. Just don't believe it's a silver bullet. It's still the same Claude. The way I use Claude is quite similar to what super powers does under the hood anyways. Like I always ask things like this: \"if I want to do X, what questi", "context": "", "use_case": "Claude Code plugin", "switched_from": "", "story": "A Rave Review of Superpowers (For Claude Code)"}, {"id": 740226464, "item_id": 46264154, "url": "https://news.ycombinator.com/item?id=46264154", "by": "radio879", "date": "2025-12-14", "stance": "recommend", "quote": "I\u2019ve been using the Superpowers plan mode: it creates a detailed plan first, then there\u2019s an execute mode", "excerpt": "Lately, I\u2019ve been using Claude Code skills/plugins. One is called Superpowers (you should be able to find it by typing \u201cclaude code superpowers github\u201d), and there are many other skills installed for all sorts of things. Gotta give props to whoever came up with the skills idea at Anthropic. It\u2019s a smart way of handling context \u2014 instead of dumping pages of helper text into the prompt (for example, how to properly bui", "context": "", "use_case": "planning and executing long coding tasks", "switched_from": "", "story": "Ask HN: How can I get better at using AI for programming?"}]}