{"slug": "amp", "name": "Amp", "category": "ai-coding", "category_name": "AI coding tools", "description": "AI coding agent from Sourcegraph for the terminal and editors.", "recommenders": 10, "commenters": 12, "warners": 0, "score": 9.553, "rising": 2.04, "first": "2025-12-08", "latest": "2026-09-15", "why": "HN commenters like Amp for solid results and fast turnaround on quick changes, along with context handoff, shareable threads and its UI.", "why_is_quote": false, "has_verdict": true, "sentiment": {"recommend": 10, "neutral": 2, "compare": 0, "warn": 0}, "spark": [0, 1, 2, 1, 0, 0, 0, 2, 0, 3, 1, 0], "listed": true, "loved": ["Result quality", "Handoff and threads"], "caveat": "Price", "short_quote": "Amp is an amazing product that I love recommending for so many reasons.", "icon": true, "domain": "ampcode.com", "github_repo": null, "official_url": "https://ampcode.com/", "pricing": "freemium", "aliases": ["Amp", "Amp Code"], "verdict": [{"text": "HN commenters like Amp for solid results and fast turnaround on quick changes,", "cites": [{"id": 744147234, "item_id": 46509202, "url": "https://news.ycombinator.com/item?id=46509202", "by": "desireco42", "date": "2026-01-05", "stance": "recommend", "quote": "Amp for a little bit that I used it, is really solid agent and delivered much better results than many others.", "excerpt": "I love Steve Yegge, but even if this thing was working how it should (and it isn't), it is just overly complicated and with a lot of misdirection. He as a dev should know that adding a layer of names on top of already named entities is not a good practice. But he just had fun and this came up. Which is fantastic. But I don't want to have to translate names in my head all the time. Just not useful. Beads also... reall", "context": "", "use_case": "coding agent", "switched_from": "", "story": "Welcome to Gas Town"}, {"id": 739193024, "item_id": 46199564, "url": "https://news.ycombinator.com/item?id=46199564", "by": "truekonrads", "date": "2025-12-08", "stance": "recommend", "quote": "I love AMP, it delivers great results.", "excerpt": "I love AMP, it delivers great results. I like that it\u2019s opinionated in how it should be used and for example tells me that I need to hand off context. I love that I am not welded to one model and someone smart has evaluated what\u2019s best fit for what. It\u2019s a bit a shame they lost Steve Yegge as their brand ambassador. A respected and practicing coder is big endorsement. To anyone on the fence - give it a go.", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Amp, Inc. \u2013 Amp is spinning out of Sourcegraph"}, {"id": 747819776, "item_id": 46738736, "url": "https://news.ycombinator.com/item?id=46738736", "by": "tecoholic", "date": "2026-01-23", "stance": "recommend", "quote": "Almost every time I need a quick change, Amp finishes the task in the time it takes Codex to build context.", "excerpt": "I use 2 cli - Codex and Amp. Almost every time I need a quick change, Amp finishes the task in the time it takes Codex to build context. I think it\u2019s got a lot to do with the system prompt and a the \u201cread loop\u201d as well, amp would read multiple files in one go and get to the task, but codex would crawl the files almost one by one. Anyone noticed this?", "context": "", "use_case": "quick code changes", "switched_from": "", "story": "Unrolling the Codex Agent Loop"}]}, {"text": " along with context handoff, shareable threads and its UI.", "cites": [{"id": 751543472, "item_id": 46971467, "url": "https://news.ycombinator.com/item?id=46971467", "by": "jauntywundrkind", "date": "2026-02-10", "stance": "recommend", "quote": "Amp is an amazing product that I love recommending for so many reasons.", "excerpt": "Fuck, that's so sad. Amp is an amazing product that I love recommending for so many reasons. Amp Free I go on and on about handoff is so so good. How great threads/sharing is, how their Oracle is just wicked good. I love being excited about how good Amp is. And there's some interest! But then I tell people there's a free (ad supported) plan, and they are signing up right away!", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Amp Free Is Full"}, {"id": 739193024, "item_id": 46199564, "url": "https://news.ycombinator.com/item?id=46199564", "by": "truekonrads", "date": "2025-12-08", "stance": "recommend", "quote": "I love AMP, it delivers great results.", "excerpt": "I love AMP, it delivers great results. I like that it\u2019s opinionated in how it should be used and for example tells me that I need to hand off context. I love that I am not welded to one model and someone smart has evaluated what\u2019s best fit for what. It\u2019s a bit a shame they lost Steve Yegge as their brand ambassador. A respected and practicing coder is big endorsement. To anyone on the fence - give it a go.", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Amp, Inc. \u2013 Amp is spinning out of Sourcegraph"}, {"id": 795414785, "item_id": 49713424, "url": "https://news.ycombinator.com/item?id=49713424", "by": "alukin", "date": "2026-09-15", "stance": "recommend", "quote": "I loved Amp Codes UI/UX and was incredibly inspired by iy", "excerpt": "Thanks for the question! I ran this a couple of times against terminal bench with different models. I saw that some models aren't as good with code mode, but overall codemode doesn't slow them down. I was usually at about 70% on tbench 2.1 with Sol. The issue I have with existing harnesses is that they try to solve all issues for all people. I have workflows I use with fork, and a history tree view is another importa", "context": "", "use_case": "coding agent", "switched_from": "", "story": "Show HN: Bough, the agent I built to replace Claude Code at work"}]}, {"text": " The main caveat raised is cost, with one fan calling it a bit expensive.", "cites": [{"id": 791801440, "item_id": 49487590, "url": "https://news.ycombinator.com/item?id=49487590", "by": "cavoirom", "date": "2026-08-29", "stance": "recommend", "quote": "AmpCode is not popular here but I recommend it", "excerpt": "AmpCode is not popular here but I recommend it, give it I try and you will not go back, except the price may a bit expensive.", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Our decision on Cursor following its acquisition by SpaceX"}]}], "praise": [{"label": "Result quality", "summary": "Commenters say it delivers better results than many other agents.", "quotes": [{"id": 744147234, "item_id": 46509202, "url": "https://news.ycombinator.com/item?id=46509202", "by": "desireco42", "date": "2026-01-05", "stance": "recommend", "quote": "Amp for a little bit that I used it, is really solid agent and delivered much better results than many others.", "excerpt": "I love Steve Yegge, but even if this thing was working how it should (and it isn't), it is just overly complicated and with a lot of misdirection. He as a dev should know that adding a layer of names on top of already named entities is not a good practice. But he just had fun and this came up. Which is fantastic. But I don't want to have to translate names in my head all the time. Just not useful. Beads also... reall", "context": "", "use_case": "coding agent", "switched_from": "", "story": "Welcome to Gas Town"}, {"id": 739193024, "item_id": 46199564, "url": "https://news.ycombinator.com/item?id=46199564", "by": "truekonrads", "date": "2025-12-08", "stance": "recommend", "quote": "I love AMP, it delivers great results.", "excerpt": "I love AMP, it delivers great results. I like that it\u2019s opinionated in how it should be used and for example tells me that I need to hand off context. I love that I am not welded to one model and someone smart has evaluated what\u2019s best fit for what. It\u2019s a bit a shame they lost Steve Yegge as their brand ambassador. A respected and practicing coder is big endorsement. To anyone on the fence - give it a go.", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Amp, Inc. \u2013 Amp is spinning out of Sourcegraph"}, {"id": 751543472, "item_id": 46971467, "url": "https://news.ycombinator.com/item?id=46971467", "by": "jauntywundrkind", "date": "2026-02-10", "stance": "recommend", "quote": "Amp is an amazing product that I love recommending for so many reasons.", "excerpt": "Fuck, that's so sad. Amp is an amazing product that I love recommending for so many reasons. Amp Free I go on and on about handoff is so so good. How great threads/sharing is, how their Oracle is just wicked good. I love being excited about how good Amp is. And there's some interest! But then I tell people there's a free (ad supported) plan, and they are signing up right away!", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Amp Free Is Full"}]}, {"label": "Handoff and threads", "summary": "Context handoff and thread sharing are singled out as strong features.", "quotes": [{"id": 751543472, "item_id": 46971467, "url": "https://news.ycombinator.com/item?id=46971467", "by": "jauntywundrkind", "date": "2026-02-10", "stance": "recommend", "quote": "Amp is an amazing product that I love recommending for so many reasons.", "excerpt": "Fuck, that's so sad. Amp is an amazing product that I love recommending for so many reasons. Amp Free I go on and on about handoff is so so good. How great threads/sharing is, how their Oracle is just wicked good. I love being excited about how good Amp is. And there's some interest! But then I tell people there's a free (ad supported) plan, and they are signing up right away!", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Amp Free Is Full"}, {"id": 739193024, "item_id": 46199564, "url": "https://news.ycombinator.com/item?id=46199564", "by": "truekonrads", "date": "2025-12-08", "stance": "recommend", "quote": "I love AMP, it delivers great results.", "excerpt": "I love AMP, it delivers great results. I like that it\u2019s opinionated in how it should be used and for example tells me that I need to hand off context. I love that I am not welded to one model and someone smart has evaluated what\u2019s best fit for what. It\u2019s a bit a shame they lost Steve Yegge as their brand ambassador. A respected and practicing coder is big endorsement. To anyone on the fence - give it a go.", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Amp, Inc. \u2013 Amp is spinning out of Sourcegraph"}]}, {"label": "Speed", "summary": "One user finds it finishes quick changes while Codex is still building context.", "quotes": [{"id": 747819776, "item_id": 46738736, "url": "https://news.ycombinator.com/item?id=46738736", "by": "tecoholic", "date": "2026-01-23", "stance": "recommend", "quote": "Almost every time I need a quick change, Amp finishes the task in the time it takes Codex to build context.", "excerpt": "I use 2 cli - Codex and Amp. Almost every time I need a quick change, Amp finishes the task in the time it takes Codex to build context. I think it\u2019s got a lot to do with the system prompt and a the \u201cread loop\u201d as well, amp would read multiple files in one go and get to the task, but codex would crawl the files almost one by one. Anyone noticed this?", "context": "", "use_case": "quick code changes", "switched_from": "", "story": "Unrolling the Codex Agent Loop"}]}, {"label": "Curated models", "summary": "A user likes not being tied to one model, with model choice evaluated for them.", "quotes": [{"id": 739193024, "item_id": 46199564, "url": "https://news.ycombinator.com/item?id=46199564", "by": "truekonrads", "date": "2025-12-08", "stance": "recommend", "quote": "I love AMP, it delivers great results.", "excerpt": "I love AMP, it delivers great results. I like that it\u2019s opinionated in how it should be used and for example tells me that I need to hand off context. I love that I am not welded to one model and someone smart has evaluated what\u2019s best fit for what. It\u2019s a bit a shame they lost Steve Yegge as their brand ambassador. A respected and practicing coder is big endorsement. To anyone on the fence - give it a go.", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Amp, Inc. \u2013 Amp is spinning out of Sourcegraph"}]}, {"label": "UI/UX", "summary": "A developer building their own harness praises its UI and UX.", "quotes": [{"id": 795414785, "item_id": 49713424, "url": "https://news.ycombinator.com/item?id=49713424", "by": "alukin", "date": "2026-09-15", "stance": "recommend", "quote": "I loved Amp Codes UI/UX and was incredibly inspired by iy", "excerpt": "Thanks for the question! I ran this a couple of times against terminal bench with different models. I saw that some models aren't as good with code mode, but overall codemode doesn't slow them down. I was usually at about 70% on tbench 2.1 with Sol. The issue I have with existing harnesses is that they try to solve all issues for all people. I have workflows I use with fork, and a history tree view is another importa", "context": "", "use_case": "coding agent", "switched_from": "", "story": "Show HN: Bough, the agent I built to replace Claude Code at work"}]}, {"label": "Value", "summary": "One subscriber says the agent alone is worth $20 a month.", "quotes": [{"id": 792030545, "item_id": 49501909, "url": "https://news.ycombinator.com/item?id=49501909", "by": "altschuler", "date": "2026-08-30", "stance": "recommend", "quote": "the Amp coding agent in itself is worth that much", "excerpt": "To give another perspective from the generally negative comments here, I've found Orbs to be just fantastic! The \"it's just a VM, everybody has that\" argument misses the point, the real value lies in just how easy it is to use and the surrounding functionality. Portals, multiplayer, local machine syncing, Puck (something like a meta assistant), referencing threads/Orbs from each other, etc. all just work as you'd hop", "context": "", "use_case": "coding agent", "switched_from": "", "story": "Orbs"}]}], "caveats": [{"label": "Price", "summary": "A recommender warns that it may be a bit expensive.", "quotes": [{"id": 791801440, "item_id": 49487590, "url": "https://news.ycombinator.com/item?id=49487590", "by": "cavoirom", "date": "2026-08-29", "stance": "recommend", "quote": "AmpCode is not popular here but I recommend it", "excerpt": "AmpCode is not popular here but I recommend it, give it I try and you will not go back, except the price may a bit expensive.", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Our decision on Cursor following its acquisition by SpaceX"}]}, {"label": "Flash for search", "summary": "One fan notes it reportedly only uses Flash for search.", "quotes": [{"id": 779470464, "item_id": 48716904, "url": "https://news.ycombinator.com/item?id=48716904", "by": "nl", "date": "2026-06-29", "stance": "recommend", "quote": "Big fan of Amp", "excerpt": "Big fan of Amp but pretty sure it only uses Flash for search: https://ampcode.com/models As for Fable: I used it as much as I could while we had it. It was a step change over Opus with my work.", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Semgrep: GLM 5.2 beats Claude in our Cyber Benchmarks"}]}, {"label": "Sandboxing advised", "summary": "A commenter recommends running agents like Amp in a sandbox, away from $HOME and SSH keys.", "quotes": [{"id": 754493154, "item_id": 47155822, "url": "https://news.ycombinator.com/item?id=47155822", "by": "downsplat", "date": "2026-02-25", "stance": "neutral", "quote": "Sandboxes are a good measure for things like Claude Code or Amp.", "excerpt": "I don't think openclaw can possibly be secured given the current paradigm. It has access to your personal stuff (that's its main use case), access to the net, and it gets untrusted third party inputs. That's the unfixable trifecta right there. No amount of filtering band-aid whack-a-mole is going to fix that. Sandboxes are a good measure for things like Claude Code or Amp. I use a bubblewrap wrapper to make sure it c", "context": "", "use_case": "coding agent, sandboxed", "switched_from": "", "story": "Sandboxes won't save you from OpenClaw"}]}], "who": [], "tags": ["CLI agent", "multi-model", "free ad-supported tier"], "use_cases": ["ai coding", "coding agent", "agent harness", "coding agents", "coding agent, sandboxed", "quick code changes"], "compared": [{"slug": "claude-code", "name": "Claude Code", "co": 4, "switched_from": 0, "switched_to": 0}, {"slug": "cursor", "name": "Cursor", "co": 3, "switched_from": 0, "switched_to": 0}, {"slug": "codex", "name": "Codex", "co": 2, "switched_from": 0, "switched_to": 0}, {"slug": "opencode", "name": "OpenCode", "co": 2, "switched_from": 0, "switched_to": 0}], "makers": [], "threads": [{"url": "https://news.ycombinator.com/item?id=49711939", "title": "Show HN: Bough, the agent I built to replace Claude Code at work", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=49437117", "title": "Orbs", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=49486172", "title": "Our decision on Cursor following its acquisition by SpaceX", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=49308073", "title": "\"That's not SoC 2 compliant\"", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=48736605", "title": "Claude Sonnet 5", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=48709670", "title": "Semgrep: GLM 5.2 beats Claude in our Cyber Benchmarks", "mentions": 1}], "sentiment_by_month": {"2025-11": {}, "2025-12": {"recommend": 1}, "2026-01": {"recommend": 2}, "2026-02": {"recommend": 1, "neutral": 1}, "2026-03": {}, "2026-04": {}, "2026-05": {}, "2026-06": {"neutral": 1, "recommend": 2}, "2026-07": {}, "2026-08": {"recommend": 3}, "2026-09": {"recommend": 1}, "2026-10": {}}, "quotes": [{"id": 795414785, "item_id": 49713424, "url": "https://news.ycombinator.com/item?id=49713424", "by": "alukin", "date": "2026-09-15", "stance": "recommend", "quote": "I loved Amp Codes UI/UX and was incredibly inspired by iy", "excerpt": "Thanks for the question! I ran this a couple of times against terminal bench with different models. I saw that some models aren't as good with code mode, but overall codemode doesn't slow them down. I was usually at about 70% on tbench 2.1 with Sol. The issue I have with existing harnesses is that they try to solve all issues for all people. I have workflows I use with fork, and a history tree view is another importa", "context": "", "use_case": "coding agent", "switched_from": "", "story": "Show HN: Bough, the agent I built to replace Claude Code at work"}, {"id": 792030545, "item_id": 49501909, "url": "https://news.ycombinator.com/item?id=49501909", "by": "altschuler", "date": "2026-08-30", "stance": "recommend", "quote": "the Amp coding agent in itself is worth that much", "excerpt": "To give another perspective from the generally negative comments here, I've found Orbs to be just fantastic! The \"it's just a VM, everybody has that\" argument misses the point, the real value lies in just how easy it is to use and the surrounding functionality. Portals, multiplayer, local machine syncing, Puck (something like a meta assistant), referencing threads/Orbs from each other, etc. all just work as you'd hop", "context": "", "use_case": "coding agent", "switched_from": "", "story": "Orbs"}, {"id": 791801440, "item_id": 49487590, "url": "https://news.ycombinator.com/item?id=49487590", "by": "cavoirom", "date": "2026-08-29", "stance": "recommend", "quote": "AmpCode is not popular here but I recommend it", "excerpt": "AmpCode is not popular here but I recommend it, give it I try and you will not go back, except the price may a bit expensive.", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Our decision on Cursor following its acquisition by SpaceX"}, {"id": 788962288, "item_id": 49310143, "url": "https://news.ycombinator.com/item?id=49310143", "by": "6ty6thhDJEHDE", "date": "2026-08-15", "stance": "recommend", "quote": "No disrespect, I love what they do.", "excerpt": "This is easy for Amp Code to do but not so easy for other people with different types of businesses. I started at a business doing millions of customer transactions worth 100s of millions of dollars and their whole process was a complete YOLO Playground. As a result they lost money, people (it is high stress if everything breaks all the time) and most importantly: customer confidence. What brought back some sanity? T", "context": "", "use_case": "", "switched_from": "", "story": "\"That's not SoC 2 compliant\""}, {"id": 779811216, "item_id": 48738201, "url": "https://news.ycombinator.com/item?id=48738201", "by": "cunningfatalist", "date": "2026-06-30", "stance": "recommend", "quote": "You might want to check out Amp: https://ampcode.com/", "excerpt": "You might want to check out Amp: https://ampcode.com/", "context": "", "use_case": "agent harness", "switched_from": "", "story": "Claude Sonnet 5"}, {"id": 779470464, "item_id": 48716904, "url": "https://news.ycombinator.com/item?id=48716904", "by": "nl", "date": "2026-06-29", "stance": "recommend", "quote": "Big fan of Amp", "excerpt": "Big fan of Amp but pretty sure it only uses Flash for search: https://ampcode.com/models As for Fable: I used it as much as I could while we had it. It was a step change over Opus with my work.", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Semgrep: GLM 5.2 beats Claude in our Cyber Benchmarks"}, {"id": 778341505, "item_id": 48646344, "url": "https://news.ycombinator.com/item?id=48646344", "by": "gls2ro", "date": "2026-06-23", "stance": "neutral", "quote": "Yes I use Claude, Coded, Amp and Cursor agents every day", "excerpt": "This is not like that. This is literally they saying they are letting their LLM run wild(ish) and seeing the status.claude.com we can see the result. This is a case where the outcome is the direct result of the engineering practices like the ones they describe. PS: Yes I use Claude, Coded, Amp and Cursor agents every day so I am not saying here LLMs are not valuable. LE: They did not made claims that \"AI is good\" the", "context": "", "use_case": "coding agents", "switched_from": "", "story": "Elevated error rate across multiple models"}, {"id": 754493154, "item_id": 47155822, "url": "https://news.ycombinator.com/item?id=47155822", "by": "downsplat", "date": "2026-02-25", "stance": "neutral", "quote": "Sandboxes are a good measure for things like Claude Code or Amp.", "excerpt": "I don't think openclaw can possibly be secured given the current paradigm. It has access to your personal stuff (that's its main use case), access to the net, and it gets untrusted third party inputs. That's the unfixable trifecta right there. No amount of filtering band-aid whack-a-mole is going to fix that. Sandboxes are a good measure for things like Claude Code or Amp. I use a bubblewrap wrapper to make sure it c", "context": "", "use_case": "coding agent, sandboxed", "switched_from": "", "story": "Sandboxes won't save you from OpenClaw"}, {"id": 751543472, "item_id": 46971467, "url": "https://news.ycombinator.com/item?id=46971467", "by": "jauntywundrkind", "date": "2026-02-10", "stance": "recommend", "quote": "Amp is an amazing product that I love recommending for so many reasons.", "excerpt": "Fuck, that's so sad. Amp is an amazing product that I love recommending for so many reasons. Amp Free I go on and on about handoff is so so good. How great threads/sharing is, how their Oracle is just wicked good. I love being excited about how good Amp is. And there's some interest! But then I tell people there's a free (ad supported) plan, and they are signing up right away!", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Amp Free Is Full"}, {"id": 747819776, "item_id": 46738736, "url": "https://news.ycombinator.com/item?id=46738736", "by": "tecoholic", "date": "2026-01-23", "stance": "recommend", "quote": "Almost every time I need a quick change, Amp finishes the task in the time it takes Codex to build context.", "excerpt": "I use 2 cli - Codex and Amp. Almost every time I need a quick change, Amp finishes the task in the time it takes Codex to build context. I think it\u2019s got a lot to do with the system prompt and a the \u201cread loop\u201d as well, amp would read multiple files in one go and get to the task, but codex would crawl the files almost one by one. Anyone noticed this?", "context": "", "use_case": "quick code changes", "switched_from": "", "story": "Unrolling the Codex Agent Loop"}, {"id": 744147234, "item_id": 46509202, "url": "https://news.ycombinator.com/item?id=46509202", "by": "desireco42", "date": "2026-01-05", "stance": "recommend", "quote": "Amp for a little bit that I used it, is really solid agent and delivered much better results than many others.", "excerpt": "I love Steve Yegge, but even if this thing was working how it should (and it isn't), it is just overly complicated and with a lot of misdirection. He as a dev should know that adding a layer of names on top of already named entities is not a good practice. But he just had fun and this came up. Which is fantastic. But I don't want to have to translate names in my head all the time. Just not useful. Beads also... reall", "context": "", "use_case": "coding agent", "switched_from": "", "story": "Welcome to Gas Town"}, {"id": 739193024, "item_id": 46199564, "url": "https://news.ycombinator.com/item?id=46199564", "by": "truekonrads", "date": "2025-12-08", "stance": "recommend", "quote": "I love AMP, it delivers great results.", "excerpt": "I love AMP, it delivers great results. I like that it\u2019s opinionated in how it should be used and for example tells me that I need to hand off context. I love that I am not welded to one model and someone smart has evaluated what\u2019s best fit for what. It\u2019s a bit a shame they lost Steve Yegge as their brand ambassador. A respected and practicing coder is big endorsement. To anyone on the fence - give it a go.", "context": "", "use_case": "AI coding", "switched_from": "", "story": "Amp, Inc. \u2013 Amp is spinning out of Sourcegraph"}]}