{
  "id": 12492840,
  "title": "Vibecoding Photoshop: Time and pressure",
  "url": "https://urgent.news/2026/10/06/vibecoding-photoshop-time-and-pressure",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-06T20:40:48.000Z",
  "source": {
    "name": "Hacker News",
    "slug": "hacker-news",
    "url": "https://andreklein.net/vibecoding-photoshop-time-and-pressure/"
  },
  "original_language": "en",
  "account": "In the world of software development, the argument that \"if anyone can now vibecode their own software, why haven't we seen any viable replacements for industry standard tools like Photoshop?\" has been debated numerous times. While it may seem that the absence of viable candidates could prove this point, two key factors often go overlooked: time and pressure. Despite ChatGPT's presence for a while, agentic coding has only recently become useful for a short time. Development still requires serious human commitment and persistence, layer by layer, debug by debug, test by test.\n\nThe more interesting question, rather than \"should this be done,\" is \"can this be done\" at all? By October 2026, the answer seems to be a resounding \"hell yeah!\" Multiple candidates have emerged, with each new tool being discovered every second day. The question of whether these tools are capable of creating a viable Photoshop alternative has been proven true by their existence alone. Photoshop serves as an excellent example for demonstrating the effectiveness of this \"agentic engineering\" approach, as its specific functionality and features can be easily verified, not just through its UI and core functionality, but also through the PSD standard.\n\nWhen evaluating these candidates, it becomes clear that they merely require more time and refinement to reach a polished state. While some issues may arise in certain esoteric functions or edge cases, this does not disprove the proof of concept. Many of these tools already appear to have gone beyond being mere proof-of-concept (POC) projects, with bugs likely to be fixed shortly after discovery.\n\nRather than focusing on nitpicky reviews of every detail, which are sure to change rapidly, it is more pragmatic to examine these projects from a broader perspective. Out of the four viable candidates presented, three are completely open-source, with a fifth expected shortly. These tools offer a range of features and forward-looking settings, such as local AI models and integrations with modern agentic harnesses. Despite some limitations, like OmaPhoto's reliance on its own .comp file format or PhotoCraft's interpretation of color adjustments, the core functionality of these tools has proven to be reliable and functional.\n\nIn conclusion, the existence of multiple viable candidates for Photoshop, all developed with the assistance of agentic coding, provides strong evidence that the initial argument regarding the unfeasibility of creating a viable Photoshop alternative has been proven incorrect. The potential for continued advancements in this field is exciting, and further testing and development will undoubtedly lead to even more impressive results in the future.",
  "summary": null,
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}