{"slug": "beads", "name": "Beads", "category": "ai-coding", "category_name": "AI coding tools", "description": "Git-backed issue tracker and memory system that coding agents use to track tasks.", "recommenders": 11, "commenters": 23, "warners": 3, "score": 8.611, "rising": 0.38, "first": "2025-11-15", "latest": "2026-10-05", "why": "HN commenters like Beads for giving coding agents a persistent task tree to plan and work from, which keeps them on track across long, multi-session projects.", "why_is_quote": false, "has_verdict": true, "sentiment": {"recommend": 10, "neutral": 10, "compare": 4, "warn": 3}, "spark": [2, 2, 3, 0, 0, 0, 0, 0, 0, 0, 3, 0], "listed": true, "loved": ["Agent continuity", "Task planning"], "caveat": "Bloat", "short_quote": "I also really like Yegge\u2019s beads for LLMs keeping state and track of what they\u2019re doing", "icon": true, "domain": null, "github_repo": "github.com/gastownhall/beads", "official_url": "https://github.com/steveyegge/beads", "pricing": "open source", "aliases": ["Beads", "beads"], "verdict": [{"text": "HN commenters like Beads for giving coding agents a persistent task tree to plan and work from", "cites": [{"id": 735011712, "item_id": 45938232, "url": "https://news.ycombinator.com/item?id=45938232", "by": "qudat", "date": "2025-11-15", "stance": "recommend", "quote": "Just use beads and you\u2019ll see what I mean.", "excerpt": "I\u2019ve done spec driven development using a bunch of markdown files. That works fine but what I have found really works is using beads: https://github.com/steveyegge/beads I\u2019m letting the agent help me draft the specs anyway and I found that the agent is a lot more focused when it can traverse a task tree using beads. It\u2019s the one spec or planning tool that I find really helps get things done without a bunch of interve", "context": "", "use_case": "agent task tracking and planning", "switched_from": "markdown spec files", "story": "Spec-Driven Development: The Waterfall Strikes Back"}, {"id": 794788800, "item_id": 49674300, "url": "https://news.ycombinator.com/item?id=49674300", "by": "djha-skin", "date": "2026-09-12", "stance": "recommend", "quote": "Using beads[1] and getting the AI to write its own tickets and wiki articles", "excerpt": "I used to think this but then improved and found that it was a matter of tooling. Using beads[1] and getting the AI to write its own tickets and wiki articles using GitHub's wiki feature on repos, it can keep track of long horizon projects pretty easily. Then you curate some of the tickets or read through them and have them update the beads or adjust them as they work them. When they want to retain some important pie", "context": "", "use_case": "long-horizon AI agent task tracking", "switched_from": "", "story": "Fuck it, make it anyway"}, {"id": 747370720, "item_id": 46710670, "url": "https://news.ycombinator.com/item?id=46710670", "by": "bravura", "date": "2026-01-21", "stance": "recommend", "quote": "We use beads for everything.", "excerpt": "We use beads for everything. We label them as \"human-spec\" needed if they are not ready to implement. We label them as \"qa-needed\" if they cannot be verified through automatic tests. I wrote beads-skills for Claude that I'll release soon to enforce this process. 2026 will be the year of agent orchestration for those of us who are frustrated having 10 different agents to check on constantly. gastown is cool but too op", "context": "", "use_case": "task tracking for agents", "switched_from": "", "story": "How are you automating your coding work?"}]}, {"text": ", which keeps them on track across long, multi-session projects.", "cites": [{"id": 747821888, "item_id": 46738868, "url": "https://news.ycombinator.com/item?id=46738868", "by": "juanre", "date": "2026-01-23", "stance": "recommend", "quote": "with beads agents don't lose track.", "excerpt": "My workflow tends to be very simple: start a session; ask the agent \"what's next\", which prompts it to check beads; and more often than not ask it to just pick up whichever bead \"makes more sense\". In claude I have a code-reviewer agent, and I remind cc often to run the code reviewer before closing any bead. It works surprisingly well. I used to monitor context and start afresh when it reached ~80%, but I stopped doi", "context": "", "use_case": "agent task tracking", "switched_from": "", "story": "Gas Town's Agent Patterns, Design Bottlenecks, and Vibecoding at Scale"}, {"id": 742871552, "item_id": 46429472, "url": "https://news.ycombinator.com/item?id=46429472", "by": "frankc", "date": "2025-12-29", "stance": "recommend", "quote": "I have been using beads for a few days on a couple of projects.", "excerpt": "Personally, I have been using beads for a few days on a couple of projects. I also like https://github.com/Dicklesworthstone/beads_viewer which is a nice tui for beads (with some additional workflow i haven't tried). I have found its been useful for longer, multi-session implementations. Its easier to get back into the work. I wouldn't go so far as to it couldn't do the work without it, but so far it seems smoother. ", "context": "", "use_case": "multi-session agent task tracking", "switched_from": "", "story": "Stop Claude Code from forgetting everything"}, {"id": 747081921, "item_id": 46692620, "url": "https://news.ycombinator.com/item?id=46692620", "by": "vessenes", "date": "2026-01-20", "stance": "recommend", "quote": "I also really like Yegge\u2019s beads for LLMs keeping state and track of what they\u2019re doing", "excerpt": "Yep, it works. Like anything getting the most out of these tools is its own (human) skill. With that in mind, a couple of comments - think of the coding agents as personalities with blind spots. A code review by all of them and a synthesis step is a good idea. In fact currently popular is the \u201crule of 5\u201d which suggests you need the LLM to review five times, and to vary the level of review, e.g. bugs, architecture, st", "context": "", "use_case": "agent state tracking", "switched_from": "", "story": "Ask HN: Do you have any evidence that agentic coding works?"}]}, {"text": " The most common complaint is that it has grown bloated and overly complicated", "cites": [{"id": 774666262, "item_id": 48416641, "url": "https://news.ycombinator.com/item?id=48416641", "by": "0xbadcafebee", "date": "2026-06-05", "stance": "warn", "quote": "at home I tried Beads but I didn't like how bloated it got", "excerpt": "1) OpenCode. Despite its many shortcomings and constant bugs, it's good enough for basic work. Use the web interface so you can switch around sessions easily, keep it running in the background (on a VPS for example). Add to it MCP servers for other systems you use, like AWS Docs MCP, Atlassian MCP, Linear MCP, SearXNG MCP, Playwright MCP, etc, which basically multiplies the AI's usefulness. Tune config to block all u", "context": "", "use_case": "AI ticketing", "switched_from": "", "story": "Ask HN: What is your (AI) dev tech stack / workflow? (June 2026)"}, {"id": 746785696, "item_id": 46674106, "url": "https://news.ycombinator.com/item?id=46674106", "by": "wild_egg", "date": "2026-01-18", "stance": "warn", "quote": "It installs magic hooks and daemons all over your system and refuses to let go. Most user hostile software I've used in a long time.", "excerpt": "Beads was phenomenal back in October when it was released. Unfortunately it has somehow grown like a cancer. Now 275k lines of Go for task tracking? And no human fully knows what it is all doing. Steve Yegge is quite proud to say he's never looked at any of its code. It installs magic hooks and daemons all over your system and refuses to let go. Most user hostile software I've used in a long time. Lot of folks rollin", "context": "", "use_case": "task tracking for AI agents", "switched_from": "", "story": "Gas Town Decoded"}, {"id": 744147233, "item_id": 46509202, "url": "https://news.ycombinator.com/item?id=46509202", "by": "desireco42", "date": "2026-01-05", "stance": "warn", "quote": "it is a task runner with labels, but it has 0 awareness of the actual tasks.", "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": "agent task tracking", "switched_from": "", "story": "Welcome to Gas Town"}]}, {"text": ", and some users have switched to simpler tools they built themselves.", "cites": [{"id": 746785696, "item_id": 46674106, "url": "https://news.ycombinator.com/item?id=46674106", "by": "wild_egg", "date": "2026-01-18", "stance": "warn", "quote": "It installs magic hooks and daemons all over your system and refuses to let go. Most user hostile software I've used in a long time.", "excerpt": "Beads was phenomenal back in October when it was released. Unfortunately it has somehow grown like a cancer. Now 275k lines of Go for task tracking? And no human fully knows what it is all doing. Steve Yegge is quite proud to say he's never looked at any of its code. It installs magic hooks and daemons all over your system and refuses to let go. Most user hostile software I've used in a long time. Lot of folks rollin", "context": "", "use_case": "task tracking for AI agents", "switched_from": "", "story": "Gas Town Decoded"}, {"id": 758815920, "item_id": 47425995, "url": "https://news.ycombinator.com/item?id=47425995", "by": "giancarlostoro", "date": "2026-03-18", "stance": "warn", "quote": "I built a similar tool out of minor frustrations with some design choices in Beads", "excerpt": "Nice, I like the UI more than mine, I built a similar tool out of minor frustrations with some design choices in Beads, mine uses SQLite exclusively instead of git or hard files, been using it for all my personal projects, but havent gone back to try and refine what I have a little more. One thing a lot of these don't do that I added to mine is synching to and from GitHub. I want people to see exactly what my local t", "context": "", "use_case": "task tracking for coding agents", "switched_from": "", "story": "Get Shit Done: A Meta-Prompting, Context Engineering and Spec-Driven Dev System"}]}], "praise": [{"label": "Agent continuity", "summary": "Agents keep track of state across sessions and context compaction, which makes it easier to resume long-running work.", "quotes": [{"id": 747821888, "item_id": 46738868, "url": "https://news.ycombinator.com/item?id=46738868", "by": "juanre", "date": "2026-01-23", "stance": "recommend", "quote": "with beads agents don't lose track.", "excerpt": "My workflow tends to be very simple: start a session; ask the agent \"what's next\", which prompts it to check beads; and more often than not ask it to just pick up whichever bead \"makes more sense\". In claude I have a code-reviewer agent, and I remind cc often to run the code reviewer before closing any bead. It works surprisingly well. I used to monitor context and start afresh when it reached ~80%, but I stopped doi", "context": "", "use_case": "agent task tracking", "switched_from": "", "story": "Gas Town's Agent Patterns, Design Bottlenecks, and Vibecoding at Scale"}, {"id": 742871552, "item_id": 46429472, "url": "https://news.ycombinator.com/item?id=46429472", "by": "frankc", "date": "2025-12-29", "stance": "recommend", "quote": "I have been using beads for a few days on a couple of projects.", "excerpt": "Personally, I have been using beads for a few days on a couple of projects. I also like https://github.com/Dicklesworthstone/beads_viewer which is a nice tui for beads (with some additional workflow i haven't tried). I have found its been useful for longer, multi-session implementations. Its easier to get back into the work. I wouldn't go so far as to it couldn't do the work without it, but so far it seems smoother. ", "context": "", "use_case": "multi-session agent task tracking", "switched_from": "", "story": "Stop Claude Code from forgetting everything"}, {"id": 747081921, "item_id": 46692620, "url": "https://news.ycombinator.com/item?id=46692620", "by": "vessenes", "date": "2026-01-20", "stance": "recommend", "quote": "I also really like Yegge\u2019s beads for LLMs keeping state and track of what they\u2019re doing", "excerpt": "Yep, it works. Like anything getting the most out of these tools is its own (human) skill. With that in mind, a couple of comments - think of the coding agents as personalities with blind spots. A code review by all of them and a synthesis step is a good idea. In fact currently popular is the \u201crule of 5\u201d which suggests you need the LLM to review five times, and to vary the level of review, e.g. bugs, architecture, st", "context": "", "use_case": "agent state tracking", "switched_from": "", "story": "Ask HN: Do you have any evidence that agentic coding works?"}]}, {"label": "Task planning", "summary": "Agents stay more focused when they can traverse a task tree and write their own tickets.", "quotes": [{"id": 735011712, "item_id": 45938232, "url": "https://news.ycombinator.com/item?id=45938232", "by": "qudat", "date": "2025-11-15", "stance": "recommend", "quote": "Just use beads and you\u2019ll see what I mean.", "excerpt": "I\u2019ve done spec driven development using a bunch of markdown files. That works fine but what I have found really works is using beads: https://github.com/steveyegge/beads I\u2019m letting the agent help me draft the specs anyway and I found that the agent is a lot more focused when it can traverse a task tree using beads. It\u2019s the one spec or planning tool that I find really helps get things done without a bunch of interve", "context": "", "use_case": "agent task tracking and planning", "switched_from": "markdown spec files", "story": "Spec-Driven Development: The Waterfall Strikes Back"}, {"id": 794788800, "item_id": 49674300, "url": "https://news.ycombinator.com/item?id=49674300", "by": "djha-skin", "date": "2026-09-12", "stance": "recommend", "quote": "Using beads[1] and getting the AI to write its own tickets and wiki articles", "excerpt": "I used to think this but then improved and found that it was a matter of tooling. Using beads[1] and getting the AI to write its own tickets and wiki articles using GitHub's wiki feature on repos, it can keep track of long horizon projects pretty easily. Then you curate some of the tickets or read through them and have them update the beads or adjust them as they work them. When they want to retain some important pie", "context": "", "use_case": "long-horizon AI agent task tracking", "switched_from": "", "story": "Fuck it, make it anyway"}, {"id": 747370720, "item_id": 46710670, "url": "https://news.ycombinator.com/item?id=46710670", "by": "bravura", "date": "2026-01-21", "stance": "recommend", "quote": "We use beads for everything.", "excerpt": "We use beads for everything. We label them as \"human-spec\" needed if they are not ready to implement. We label them as \"qa-needed\" if they cannot be verified through automatic tests. I wrote beads-skills for Claude that I'll release soon to enforce this process. 2026 will be the year of agent orchestration for those of us who are frustrated having 10 different agents to check on constantly. gastown is cool but too op", "context": "", "use_case": "task tracking for agents", "switched_from": "", "story": "How are you automating your coding work?"}]}, {"label": "Data architecture", "summary": "Users like the JSONL files checked into git, backed by a local SQLite cache.", "quotes": [{"id": 737293264, "item_id": 46080829, "url": "https://news.ycombinator.com/item?id=46080829", "by": "jauntywundrkind", "date": "2025-11-28", "stance": "recommend", "quote": "Somewhat aside but I love the data architrcture.", "excerpt": "Somewhat aside but I love the data architrcture. Jsonl lines checked into git, and a sqlite local cache. Auto-sync to replicate changes between the canonical and cache. https://github.com/steveyegge/beads?tab=readme-ov-file#the-magic-distributed-database-via-git I finally started digging in to OpenCode for real these past couple weeks. It has a planning mode, which nicely builds out a plan on text chat as usual, but ", "context": "", "use_case": "agent memory and task tracking", "switched_from": "", "story": "Beads \u2013 A memory upgrade for your coding agent"}, {"id": 746784512, "item_id": 46674032, "url": "https://news.ycombinator.com/item?id=46674032", "by": "fdr", "date": "2026-01-18", "stance": "recommend", "quote": "I use beads quite a bit, but not as steve intended.", "excerpt": "I use beads quite a bit, but not as steve intended. And definitely the opposite of \"Gas Town,\" where I use the note-taking capability and integration with Git (that is, as something of a glorified Makefile and database) to debug contexts, to close the loop and increase accuracy over time. Nevertheless, it has been useful for large batch runs over my code base: the record has been processing for thirty hours straight ", "context": "", "use_case": "debugging agent contexts, batch runs", "switched_from": "", "story": "Gas Town Decoded"}]}, {"label": "Decision records", "summary": "Some users store notes and design decisions in beads to debug contexts and coordinate agents.", "quotes": [{"id": 799341378, "item_id": 49958836, "url": "https://news.ycombinator.com/item?id=49958836", "by": "mtzaldo", "date": "2026-10-04", "stance": "neutral", "quote": "beads store decisions as well (including design ones), then multiple agents work on a single bead", "excerpt": "in my case, i use opencode with oh-my-openagents. Omo has an orchestrator in charge and manage all the intetactions between agents. ADRs store decisions and beads store decisions as well (including design ones), then multiple agents work on a single bead.", "context": "", "use_case": "storing agent decisions and tasks", "switched_from": "", "story": "We want you to build the next Git platform on Cloudflare"}, {"id": 746784512, "item_id": 46674032, "url": "https://news.ycombinator.com/item?id=46674032", "by": "fdr", "date": "2026-01-18", "stance": "recommend", "quote": "I use beads quite a bit, but not as steve intended.", "excerpt": "I use beads quite a bit, but not as steve intended. And definitely the opposite of \"Gas Town,\" where I use the note-taking capability and integration with Git (that is, as something of a glorified Makefile and database) to debug contexts, to close the loop and increase accuracy over time. Nevertheless, it has been useful for large batch runs over my code base: the record has been processing for thirty hours straight ", "context": "", "use_case": "debugging agent contexts, batch runs", "switched_from": "", "story": "Gas Town Decoded"}]}, {"label": "Single-agent focus", "summary": "One user prefers optimizing a single agent with Beads over running many agents.", "quotes": [{"id": 751897568, "item_id": 46993598, "url": "https://news.ycombinator.com/item?id=46993598", "by": "giancarlostoro", "date": "2026-02-12", "stance": "recommend", "quote": "I'm in the latter, I don't have infinite resources, I'd rather stick to one agent and optimize what it can do.", "excerpt": "I feel like there's two camps: * Throw more agents * Use something like Beads I'm in the latter, I don't have infinite resources, I'd rather stick to one agent and optimize what it can do. When I hit my Claude Code limit, I stop, I use Claude Code primarily for side projects.", "context": "limited resources, side projects", "use_case": "single-agent workflow", "switched_from": "", "story": "Show HN: 20+ Claude Code agents coordinating on real work (open source)"}]}], "caveats": [{"label": "Bloat", "summary": "Users say it has grown large and overly complicated compared with its early versions.", "quotes": [{"id": 746785696, "item_id": 46674106, "url": "https://news.ycombinator.com/item?id=46674106", "by": "wild_egg", "date": "2026-01-18", "stance": "warn", "quote": "It installs magic hooks and daemons all over your system and refuses to let go. Most user hostile software I've used in a long time.", "excerpt": "Beads was phenomenal back in October when it was released. Unfortunately it has somehow grown like a cancer. Now 275k lines of Go for task tracking? And no human fully knows what it is all doing. Steve Yegge is quite proud to say he's never looked at any of its code. It installs magic hooks and daemons all over your system and refuses to let go. Most user hostile software I've used in a long time. Lot of folks rollin", "context": "", "use_case": "task tracking for AI agents", "switched_from": "", "story": "Gas Town Decoded"}, {"id": 774666262, "item_id": 48416641, "url": "https://news.ycombinator.com/item?id=48416641", "by": "0xbadcafebee", "date": "2026-06-05", "stance": "warn", "quote": "at home I tried Beads but I didn't like how bloated it got", "excerpt": "1) OpenCode. Despite its many shortcomings and constant bugs, it's good enough for basic work. Use the web interface so you can switch around sessions easily, keep it running in the background (on a VPS for example). Add to it MCP servers for other systems you use, like AWS Docs MCP, Atlassian MCP, Linear MCP, SearXNG MCP, Playwright MCP, etc, which basically multiplies the AI's usefulness. Tune config to block all u", "context": "", "use_case": "AI ticketing", "switched_from": "", "story": "Ask HN: What is your (AI) dev tech stack / workflow? (June 2026)"}, {"id": 744147233, "item_id": 46509202, "url": "https://news.ycombinator.com/item?id=46509202", "by": "desireco42", "date": "2026-01-05", "stance": "warn", "quote": "it is a task runner with labels, but it has 0 awareness of the actual tasks.", "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": "agent task tracking", "switched_from": "", "story": "Welcome to Gas Town"}]}, {"label": "Intrusive install", "summary": "One user says it installs hooks and daemons across the system that are hard to remove.", "quotes": [{"id": 746785696, "item_id": 46674106, "url": "https://news.ycombinator.com/item?id=46674106", "by": "wild_egg", "date": "2026-01-18", "stance": "warn", "quote": "It installs magic hooks and daemons all over your system and refuses to let go. Most user hostile software I've used in a long time.", "excerpt": "Beads was phenomenal back in October when it was released. Unfortunately it has somehow grown like a cancer. Now 275k lines of Go for task tracking? And no human fully knows what it is all doing. Steve Yegge is quite proud to say he's never looked at any of its code. It installs magic hooks and daemons all over your system and refuses to let go. Most user hostile software I've used in a long time. Lot of folks rollin", "context": "", "use_case": "task tracking for AI agents", "switched_from": "", "story": "Gas Town Decoded"}]}, {"label": "Simpler alternatives", "summary": "Some users found custom tools, plain worklogs or Fossil worked as well or better.", "quotes": [{"id": 758815920, "item_id": 47425995, "url": "https://news.ycombinator.com/item?id=47425995", "by": "giancarlostoro", "date": "2026-03-18", "stance": "warn", "quote": "I built a similar tool out of minor frustrations with some design choices in Beads", "excerpt": "Nice, I like the UI more than mine, I built a similar tool out of minor frustrations with some design choices in Beads, mine uses SQLite exclusively instead of git or hard files, been using it for all my personal projects, but havent gone back to try and refine what I have a little more. One thing a lot of these don't do that I added to mine is synching to and from GitHub. I want people to see exactly what my local t", "context": "", "use_case": "task tracking for coding agents", "switched_from": "", "story": "Get Shit Done: A Meta-Prompting, Context Engineering and Spec-Driven Dev System"}, {"id": 799198465, "item_id": 49949904, "url": "https://news.ycombinator.com/item?id=49949904", "by": "zhynn", "date": "2026-10-04", "stance": "compare", "quote": "I looked into beads but I didn't see a way to make it work any better than fossil, and fossil does more stuff.", "excerpt": "I looked into beads but I didn't see a way to make it work any better than fossil, and fossil does more stuff. My typical setup is to use the repo as a working directory, the tickets for todo, wiki/forum for conversation. There is also a chat interface, but I haven't tried that with agents yet. But I love fossil for this. You can create users for all of the agents (or agent roles). The CLI for fossil is just a binary", "context": "", "use_case": "agent work tracking", "switched_from": "", "story": "We want you to build the next Git platform on Cloudflare"}, {"id": 797711264, "item_id": 49856954, "url": "https://news.ycombinator.com/item?id=49856954", "by": "Huppie", "date": "2026-09-26", "stance": "recommend", "quote": "I still run an old version without dolt that's imho pretty good overall", "excerpt": "Not OP but will definitely have a look. I have some older projects that use beads (I still run an old version without dolt that's imho pretty good overall) but lately with Fable also have a few newer projects where I just have the agent write docs and keep a worklog with the what/why/decisions etc. (I think I read it here on HN somewhere and figured I'd give that a try.) The latter seems to work pretty well for now (", "context": "", "use_case": "agent task state tracking", "switched_from": "", "story": "Plan mode is dead"}]}, {"label": "Solo workflows", "summary": "One solo, single-threaded user found it got in the way and sees more value with many simultaneous agents.", "quotes": [{"id": 793931024, "item_id": 49620689, "url": "https://news.ycombinator.com/item?id=49620689", "by": "oefrha", "date": "2026-09-09", "stance": "neutral", "quote": "it didn\u2019t end up being a force multiplier for my workflows (and actually kinda got in the way from time to time due to me rebasing too much), but I can see its value", "excerpt": "Sounds like https://github.com/gastownhall/beads to me. I prefer working solo, single-threaded on components to stay on top of agentic work, so it didn\u2019t end up being a force multiplier for my workflows (and actually kinda got in the way from time to time due to me rebasing too much), but I can see its value when you have many agents working simultaneously.", "context": "prefers working solo, single-threaded on components", "use_case": "agent task tracking", "switched_from": "", "story": "An Accidental Blackboard"}]}, {"label": "Rapid churn", "summary": "Commenters describe it as fast-changing and needing time to settle, and one calls it vibe-coded.", "quotes": [{"id": 744105825, "item_id": 46506614, "url": "https://news.ycombinator.com/item?id=46506614", "by": "vessenes", "date": "2026-01-05", "stance": "neutral", "quote": "if beads is any indication, give it a month or two or four to settle down", "excerpt": "I put in 15 hours or so with gas town this weekend, from just around the 0.1 release. Think of as an extended bipolar-optimism-fueled glimpse into the future. Steve's MO is laid out in the medium post - but basically, it's okay to lose things, rewrite whole subsystems, whatever, this is the future. It's really fun and interesting to watch the speed of development. I've made a few multi agent coding setups in the last", "context": "", "use_case": "", "switched_from": "", "story": "Welcome to Gas Town"}, {"id": 787016816, "item_id": 49188551, "url": "https://news.ycombinator.com/item?id=49188551", "by": "CharlesW", "date": "2026-08-05", "stance": "compare", "quote": "So is this basically Beads + Dolt?", "excerpt": "So is this basically Beads + Dolt? This isn\u2019t to diminish it, it\u2019d be great if this ends up being a less \u201cvibe code-y\u201d alternative to those.", "context": "", "use_case": "", "switched_from": "", "story": "Zed DeltaDB"}]}, {"label": "Limited awareness", "summary": "One critic says it is a task runner with labels and no awareness of the actual tasks.", "quotes": [{"id": 744147233, "item_id": 46509202, "url": "https://news.ycombinator.com/item?id=46509202", "by": "desireco42", "date": "2026-01-05", "stance": "warn", "quote": "it is a task runner with labels, but it has 0 awareness of the actual tasks.", "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": "agent task tracking", "switched_from": "", "story": "Welcome to Gas Town"}]}], "who": [{"by": "oefrha", "context": "prefers working solo, single-threaded on components", "url": "https://news.ycombinator.com/item?id=49620689", "date": "2026-09-09"}, {"by": "giancarlostoro", "context": "limited resources, side projects", "url": "https://news.ycombinator.com/item?id=46993598", "date": "2026-02-12"}], "tags": ["for Claude Code users", "for multi-session projects", "for multi-agent setups", "git-backed"], "use_cases": ["agent task tracking", "task tracking for agents", "coding agent task tracking", "storing agent decisions and tasks", "agent work tracking", "agent task state tracking"], "compared": [{"slug": "claude-code", "name": "Claude Code", "co": 9, "switched_from": 0, "switched_to": 0}, {"slug": "gas-town", "name": "Gas Town", "co": 5, "switched_from": 0, "switched_to": 0}, {"slug": "claude", "name": "Claude", "co": 4, "switched_from": 0, "switched_to": 0}, {"slug": "codex", "name": "Codex", "co": 3, "switched_from": 0, "switched_to": 0}, {"slug": "opencode", "name": "OpenCode", "co": 3, "switched_from": 0, "switched_to": 0}], "makers": [], "threads": [{"url": "https://news.ycombinator.com/item?id=49947051", "title": "We want you to build the next Git platform on Cloudflare", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=46624883", "title": "Gas Town Decoded", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=46458936", "title": "Welcome to Gas Town", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=46075616", "title": "Beads \u2013 A memory upgrade for your coding agent", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=49945933", "title": "Agents don't need memory, they need documentation", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=49840054", "title": "Plan mode is dead", "mentions": 1}], "sentiment_by_month": {"2025-11": {"recommend": 2, "neutral": 1}, "2025-12": {"recommend": 2}, "2026-01": {"neutral": 1, "warn": 2, "compare": 1, "recommend": 3}, "2026-02": {"compare": 1}, "2026-03": {"neutral": 1}, "2026-04": {"neutral": 2}, "2026-05": {"neutral": 1}, "2026-06": {"warn": 1, "neutral": 1}, "2026-07": {}, "2026-08": {"compare": 1}, "2026-09": {"neutral": 1, "recommend": 3}, "2026-10": {"compare": 1, "neutral": 2}}, "quotes": [{"id": 799372466, "item_id": 49960779, "url": "https://news.ycombinator.com/item?id=49960779", "by": "kaydub", "date": "2026-10-05", "stance": "neutral", "quote": "I DO still use beads on a lot of projects, but not always.", "excerpt": "Nope. Opus and Sol for the most part. Haven't even used Astra or Fable. At work it's all Anthropic. I spent a while on Sonnet models but have been pretty consistent about using Opus now. For personal projects I bounce between Sol/Terra and Gemini Flash. Most of the models are good enough now. You don't need all these documents, memory, plugins, skills, etc. MCPs are good for hooking up to external sources of data (at", "context": "", "use_case": "agent task tracking", "switched_from": "", "story": "Agents don't need memory, they need documentation"}, {"id": 799341378, "item_id": 49958836, "url": "https://news.ycombinator.com/item?id=49958836", "by": "mtzaldo", "date": "2026-10-04", "stance": "neutral", "quote": "beads store decisions as well (including design ones), then multiple agents work on a single bead", "excerpt": "in my case, i use opencode with oh-my-openagents. Omo has an orchestrator in charge and manage all the intetactions between agents. ADRs store decisions and beads store decisions as well (including design ones), then multiple agents work on a single bead.", "context": "", "use_case": "storing agent decisions and tasks", "switched_from": "", "story": "We want you to build the next Git platform on Cloudflare"}, {"id": 799198465, "item_id": 49949904, "url": "https://news.ycombinator.com/item?id=49949904", "by": "zhynn", "date": "2026-10-04", "stance": "compare", "quote": "I looked into beads but I didn't see a way to make it work any better than fossil, and fossil does more stuff.", "excerpt": "I looked into beads but I didn't see a way to make it work any better than fossil, and fossil does more stuff. My typical setup is to use the repo as a working directory, the tickets for todo, wiki/forum for conversation. There is also a chat interface, but I haven't tried that with agents yet. But I love fossil for this. You can create users for all of the agents (or agent roles). The CLI for fossil is just a binary", "context": "", "use_case": "agent work tracking", "switched_from": "", "story": "We want you to build the next Git platform on Cloudflare"}, {"id": 797711264, "item_id": 49856954, "url": "https://news.ycombinator.com/item?id=49856954", "by": "Huppie", "date": "2026-09-26", "stance": "recommend", "quote": "I still run an old version without dolt that's imho pretty good overall", "excerpt": "Not OP but will definitely have a look. I have some older projects that use beads (I still run an old version without dolt that's imho pretty good overall) but lately with Fable also have a few newer projects where I just have the agent write docs and keep a worklog with the what/why/decisions etc. (I think I read it here on HN somewhere and figured I'd give that a try.) The latter seems to work pretty well for now (", "context": "", "use_case": "agent task state tracking", "switched_from": "", "story": "Plan mode is dead"}, {"id": 796944928, "item_id": 49809058, "url": "https://news.ycombinator.com/item?id=49809058", "by": "sroerick", "date": "2026-09-22", "stance": "recommend", "quote": "I also think Yegge's beads and Gastown have some really smart things to say in this regard", "excerpt": "Personally, I build a loosely compiled / linted DSL to handle spec. It's a SEXP language but could just as easily be markdown. It'd probably be better if it was - I spend a lot of tokens matching parentheses. Compared to markdown - I lose some human readability but gain a lot in workflow. For me - it turns the sort of \"free form jazz code odyssey\" into something with structure. It changes the question to \"does the sp", "context": "", "use_case": "agent spec/workflow", "switched_from": "", "story": "Markdown in /src"}, {"id": 794788800, "item_id": 49674300, "url": "https://news.ycombinator.com/item?id=49674300", "by": "djha-skin", "date": "2026-09-12", "stance": "recommend", "quote": "Using beads[1] and getting the AI to write its own tickets and wiki articles", "excerpt": "I used to think this but then improved and found that it was a matter of tooling. Using beads[1] and getting the AI to write its own tickets and wiki articles using GitHub's wiki feature on repos, it can keep track of long horizon projects pretty easily. Then you curate some of the tickets or read through them and have them update the beads or adjust them as they work them. When they want to retain some important pie", "context": "", "use_case": "long-horizon AI agent task tracking", "switched_from": "", "story": "Fuck it, make it anyway"}, {"id": 793931024, "item_id": 49620689, "url": "https://news.ycombinator.com/item?id=49620689", "by": "oefrha", "date": "2026-09-09", "stance": "neutral", "quote": "it didn\u2019t end up being a force multiplier for my workflows (and actually kinda got in the way from time to time due to me rebasing too much), but I can see its value", "excerpt": "Sounds like https://github.com/gastownhall/beads to me. I prefer working solo, single-threaded on components to stay on top of agentic work, so it didn\u2019t end up being a force multiplier for my workflows (and actually kinda got in the way from time to time due to me rebasing too much), but I can see its value when you have many agents working simultaneously.", "context": "prefers working solo, single-threaded on components", "use_case": "agent task tracking", "switched_from": "", "story": "An Accidental Blackboard"}, {"id": 787016816, "item_id": 49188551, "url": "https://news.ycombinator.com/item?id=49188551", "by": "CharlesW", "date": "2026-08-05", "stance": "compare", "quote": "So is this basically Beads + Dolt?", "excerpt": "So is this basically Beads + Dolt? This isn\u2019t to diminish it, it\u2019d be great if this ends up being a less \u201cvibe code-y\u201d alternative to those.", "context": "", "use_case": "", "switched_from": "", "story": "Zed DeltaDB"}, {"id": 776326963, "item_id": 48520435, "url": "https://news.ycombinator.com/item?id=48520435", "by": "jessepcc", "date": "2026-06-13", "stance": "neutral", "quote": "Planned to building on beads to for non-repo based tasks", "excerpt": "Great project, will give it a try! Struggling to find a simple task management for work across different repos and some admin+marketing tasks. Become very messy when tasks are distributed among Hermes, Claude Code & myself. Planned to building on beads to for non-repo based tasks, adding long term goal to keep driving agents busy, UI mainly sits on telegram group with topics. See whether Paca can solve the problem :)", "context": "", "use_case": "task tracking for agents", "switched_from": "", "story": "Show HN: Paca \u2013 Lightweight Jira alternative for human-AI collaboration"}, {"id": 774666262, "item_id": 48416641, "url": "https://news.ycombinator.com/item?id=48416641", "by": "0xbadcafebee", "date": "2026-06-05", "stance": "warn", "quote": "at home I tried Beads but I didn't like how bloated it got", "excerpt": "1) OpenCode. Despite its many shortcomings and constant bugs, it's good enough for basic work. Use the web interface so you can switch around sessions easily, keep it running in the background (on a VPS for example). Add to it MCP servers for other systems you use, like AWS Docs MCP, Atlassian MCP, Linear MCP, SearXNG MCP, Playwright MCP, etc, which basically multiplies the AI's usefulness. Tune config to block all u", "context": "", "use_case": "AI ticketing", "switched_from": "", "story": "Ask HN: What is your (AI) dev tech stack / workflow? (June 2026)"}, {"id": 771322545, "item_id": 48207659, "url": "https://news.ycombinator.com/item?id=48207659", "by": "giancarlostoro", "date": "2026-05-20", "stance": "neutral", "quote": "I used to use Beads before I made GuardRails", "excerpt": "This is precisely how I used to use Beads before I made GuardRails (I wanted something slightly simpler, but similar with more 'guard rails'). I braindump everything I want to build, I ask Claude to do market level research. I then ask Claude to ask clarifying questions, when I ask Claude to be critical of its conclusions and provide the top options and to justify it. I also question Claude and say its okay to disagr", "context": "", "use_case": "task tracking for agents", "switched_from": "", "story": "Learnings from 100K lines of Rust with AI (2025)"}, {"id": 764605267, "item_id": 47787829, "url": "https://news.ycombinator.com/item?id=47787829", "by": "headcanon", "date": "2026-04-15", "stance": "neutral", "quote": "currently beads in my main 'wiki' monorepo", "excerpt": "I tried openclaw when it was released, but I preferred the minimalism of nanoclaw and replaced it. I have it on a mac mini now For context: I have additional automation with scripts set up on the mini, some of them call LLMs to do things like summarizing today's news. I have other automations that are agentic and just run \"claude -p\" (mainly just checks status of other jobs and fixes them automatically). Agentic auto", "context": "", "use_case": "centralized task list", "switched_from": "", "story": "Is anyone actually using OpenClaw?"}, {"id": 764332704, "item_id": 47770794, "url": "https://news.ycombinator.com/item?id=47770794", "by": "giancarlostoro", "date": "2026-04-14", "stance": "neutral", "quote": "I loved Beads, but kept running into issues because it is so git heavy.", "excerpt": "I loved Beads, but kept running into issues because it is so git heavy. One: not every system and project I work on uses git. Two: Sometimes I'd switch branches, and that would screw up Beads state entirely. Three: And this is at least last I used it, there's no safety net, Claude would close a Bead, without validating anything. I wound up building my own with Claude, I made it SQLite first, syncs to GitHub, can pull", "context": "", "use_case": "agent task tracking", "switched_from": "", "story": "Gas Town: From Clown Show to v1.0"}, {"id": 758815920, "item_id": 47425995, "url": "https://news.ycombinator.com/item?id=47425995", "by": "giancarlostoro", "date": "2026-03-18", "stance": "warn", "quote": "I built a similar tool out of minor frustrations with some design choices in Beads", "excerpt": "Nice, I like the UI more than mine, I built a similar tool out of minor frustrations with some design choices in Beads, mine uses SQLite exclusively instead of git or hard files, been using it for all my personal projects, but havent gone back to try and refine what I have a little more. One thing a lot of these don't do that I added to mine is synching to and from GitHub. I want people to see exactly what my local t", "context": "", "use_case": "task tracking for coding agents", "switched_from": "", "story": "Get Shit Done: A Meta-Prompting, Context Engineering and Spec-Driven Dev System"}, {"id": 756628721, "item_id": 47289295, "url": "https://news.ycombinator.com/item?id=47289295", "by": "giancarlostoro", "date": "2026-03-07", "stance": "neutral", "quote": "I used to use Beads and now GuardRails", "excerpt": "This is why I used to use Beads and now GuardRails (shameless plug[0]). You brain dump to the model what you want, it breaks it down into discrete tasks, you have it refine them with you. By the time you have the model work on everything it can spawn workers in parallel that know what to do. In hindsight I should have called it BrainDump. [0]: https://giancarlostoro.com/introducing-guardrails-a-new-coding-agent-task-", "context": "", "use_case": "coding agent task tracking", "switched_from": "", "story": "LLM Doesn't Write Correct Code. It Writes Plausible Code"}, {"id": 755569169, "item_id": 47223073, "url": "https://news.ycombinator.com/item?id=47223073", "by": "giancarlostoro", "date": "2026-03-02", "stance": "neutral", "quote": "This is how I used to use Beads before I made GuardRails[0].", "excerpt": "This is how I used to use Beads before I made GuardRails[0]. I basically iterate with the model, ask it to do market research, review everything it suggests, and you wind up with a \"prompt\" that tells it what to do and how to work that was designed by the model using its own known verbiage. Having learned about how XML could be used to influence Claude I'm rethinking my flow and how GuardRails behaves. [0]: https://g", "context": "", "use_case": "agent task tracking", "switched_from": "", "story": "If AI writes code, should the session be part of the commit?"}, {"id": 754466241, "item_id": 47154140, "url": "https://news.ycombinator.com/item?id=47154140", "by": "giancarlostoro", "date": "2026-02-25", "stance": "compare", "quote": "Beads is a behemoth of code", "excerpt": "> Instead of extensions you install, you download a skill file that tells a coding agent how to add a feature. The software stops being an artifact and starts being a living tool that isn't the same as anyone else's copy. I'm curious to see what tooling will emerge for collaborating with this new paradigm. I build my own inspired by Beads, not quite as you're describing, but I store todo's in a SQLite database (beads", "context": "", "use_case": "agent task tracking", "switched_from": "", "story": "Pi \u2013 a minimal terminal coding harness"}, {"id": 752658065, "item_id": 47041129, "url": "https://news.ycombinator.com/item?id=47041129", "by": "giancarlostoro", "date": "2026-02-16", "stance": "compare", "quote": "I didnt like how married to git hooks beads was", "excerpt": "I'm aware I've been building my own alternative to Beads for weeks now. the instructions.md file and something like Beads is epic. https://github.com/Giancarlos/guardrails I didnt like how married to git hooks beads was, so I made my own thats primarily a SQLite workhorse. Been using it just the same as I have used Beads, works just as good, drastically less code. I added a concept called \"gates\" to stop the AI model", "context": "", "use_case": "AI agent task tracking", "switched_from": "", "story": "OpenClaw, OpenAI and the Future"}, {"id": 751897568, "item_id": 46993598, "url": "https://news.ycombinator.com/item?id=46993598", "by": "giancarlostoro", "date": "2026-02-12", "stance": "recommend", "quote": "I'm in the latter, I don't have infinite resources, I'd rather stick to one agent and optimize what it can do.", "excerpt": "I feel like there's two camps: * Throw more agents * Use something like Beads I'm in the latter, I don't have infinite resources, I'd rather stick to one agent and optimize what it can do. When I hit my Claude Code limit, I stop, I use Claude Code primarily for side projects.", "context": "limited resources, side projects", "use_case": "single-agent workflow", "switched_from": "", "story": "Show HN: 20+ Claude Code agents coordinating on real work (open source)"}, {"id": 751442288, "item_id": 46965143, "url": "https://news.ycombinator.com/item?id=46965143", "by": "giancarlostoro", "date": "2026-02-10", "stance": "compare", "quote": "I've been building my own alternative to Beads", "excerpt": "I'll be sure to try these out. I've been building my own alternative to Beads with a concept called \"gates\" which do not let you close tasks as complete until a gate passes. Would love to throw these in as \"gates\" for my current workflow.", "context": "building their own alternative", "use_case": "agent task tracking", "switched_from": "", "story": "Show HN: Showboat and Rodney, so agents can demo what they've built"}, {"id": 751390561, "item_id": 46961910, "url": "https://news.ycombinator.com/item?id=46961910", "by": "giancarlostoro", "date": "2026-02-10", "stance": "recommend", "quote": "Funnily enough, I love Beads, but did not like that it uses git hooks for the DB", "excerpt": "> Spec-driven development is becoming the primary driver of code generation. This sounds like my current \"phase\" of AI coding. I have had so many project ideas for years that I can just spec out, everything I've thought about, all the little ideas and details, things I only had time to think about, never implement. I then feed it to Claude, and watch it meet my every specification, I can then test it, note any bugs, ", "context": "", "use_case": "agent ticket tracking", "switched_from": "", "story": "Ex-GitHub CEO Launches a New Developer Platform for AI Agents"}, {"id": 750474016, "item_id": 46904626, "url": "https://news.ycombinator.com/item?id=46904626", "by": "giancarlostoro", "date": "2026-02-05", "stance": "compare", "quote": "I was working on my own alternative to Beads", "excerpt": "I was working on my own alternative to Beads... then I realized I could do exactly this with something similar to Beads, I'm planning on open sourcing it soon because I like what I have so far, I also made it so I can sync my tasks directly to my GitHub projects as well. I think its more useful to have agent tasks eventually synched back up to real ticketing systems for historical reasons. Besides, its better to have", "context": "building own agent task tool", "use_case": "agent task tracking", "switched_from": "", "story": "Claude Code Agent Teams"}, {"id": 747821888, "item_id": 46738868, "url": "https://news.ycombinator.com/item?id=46738868", "by": "juanre", "date": "2026-01-23", "stance": "recommend", "quote": "with beads agents don't lose track.", "excerpt": "My workflow tends to be very simple: start a session; ask the agent \"what's next\", which prompts it to check beads; and more often than not ask it to just pick up whichever bead \"makes more sense\". In claude I have a code-reviewer agent, and I remind cc often to run the code reviewer before closing any bead. It works surprisingly well. I used to monitor context and start afresh when it reached ~80%, but I stopped doi", "context": "", "use_case": "agent task tracking", "switched_from": "", "story": "Gas Town's Agent Patterns, Design Bottlenecks, and Vibecoding at Scale"}, {"id": 747370720, "item_id": 46710670, "url": "https://news.ycombinator.com/item?id=46710670", "by": "bravura", "date": "2026-01-21", "stance": "recommend", "quote": "We use beads for everything.", "excerpt": "We use beads for everything. We label them as \"human-spec\" needed if they are not ready to implement. We label them as \"qa-needed\" if they cannot be verified through automatic tests. I wrote beads-skills for Claude that I'll release soon to enforce this process. 2026 will be the year of agent orchestration for those of us who are frustrated having 10 different agents to check on constantly. gastown is cool but too op", "context": "", "use_case": "task tracking for agents", "switched_from": "", "story": "How are you automating your coding work?"}, {"id": 747081921, "item_id": 46692620, "url": "https://news.ycombinator.com/item?id=46692620", "by": "vessenes", "date": "2026-01-20", "stance": "recommend", "quote": "I also really like Yegge\u2019s beads for LLMs keeping state and track of what they\u2019re doing", "excerpt": "Yep, it works. Like anything getting the most out of these tools is its own (human) skill. With that in mind, a couple of comments - think of the coding agents as personalities with blind spots. A code review by all of them and a synthesis step is a good idea. In fact currently popular is the \u201crule of 5\u201d which suggests you need the LLM to review five times, and to vary the level of review, e.g. bugs, architecture, st", "context": "", "use_case": "agent state tracking", "switched_from": "", "story": "Ask HN: Do you have any evidence that agentic coding works?"}, {"id": 746785696, "item_id": 46674106, "url": "https://news.ycombinator.com/item?id=46674106", "by": "wild_egg", "date": "2026-01-18", "stance": "warn", "quote": "It installs magic hooks and daemons all over your system and refuses to let go. Most user hostile software I've used in a long time.", "excerpt": "Beads was phenomenal back in October when it was released. Unfortunately it has somehow grown like a cancer. Now 275k lines of Go for task tracking? And no human fully knows what it is all doing. Steve Yegge is quite proud to say he's never looked at any of its code. It installs magic hooks and daemons all over your system and refuses to let go. Most user hostile software I've used in a long time. Lot of folks rollin", "context": "", "use_case": "task tracking for AI agents", "switched_from": "", "story": "Gas Town Decoded"}, {"id": 746784512, "item_id": 46674032, "url": "https://news.ycombinator.com/item?id=46674032", "by": "fdr", "date": "2026-01-18", "stance": "recommend", "quote": "I use beads quite a bit, but not as steve intended.", "excerpt": "I use beads quite a bit, but not as steve intended. And definitely the opposite of \"Gas Town,\" where I use the note-taking capability and integration with Git (that is, as something of a glorified Makefile and database) to debug contexts, to close the loop and increase accuracy over time. Nevertheless, it has been useful for large batch runs over my code base: the record has been processing for thirty hours straight ", "context": "", "use_case": "debugging agent contexts, batch runs", "switched_from": "", "story": "Gas Town Decoded"}, {"id": 744159120, "item_id": 46509945, "url": "https://news.ycombinator.com/item?id=46509945", "by": "bredren", "date": "2026-01-06", "stance": "compare", "quote": "I have been looking at this problem space, including beads recently.", "excerpt": "I have been looking at this problem space, including beads recently. I wasn't able to find something AI-native that supported subissues, and worked across projects. I require git worktrees on my programming work now, and I want it to be able to handle integration with systems like Linear or even GH issues (which was my preferred ticketing system before CLIs emerged) The idea is in part, ~I have 15 minutes right now, ", "context": "", "use_case": "AI task tracking", "switched_from": "", "story": "Show HN: I replaced beads with a faster, simpler Markdown-based task tracker"}, {"id": 744147233, "item_id": 46509202, "url": "https://news.ycombinator.com/item?id=46509202", "by": "desireco42", "date": "2026-01-05", "stance": "warn", "quote": "it is a task runner with labels, but it has 0 awareness of the actual tasks.", "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": "agent task tracking", "switched_from": "", "story": "Welcome to Gas Town"}, {"id": 744105825, "item_id": 46506614, "url": "https://news.ycombinator.com/item?id=46506614", "by": "vessenes", "date": "2026-01-05", "stance": "neutral", "quote": "if beads is any indication, give it a month or two or four to settle down", "excerpt": "I put in 15 hours or so with gas town this weekend, from just around the 0.1 release. Think of as an extended bipolar-optimism-fueled glimpse into the future. Steve's MO is laid out in the medium post - but basically, it's okay to lose things, rewrite whole subsystems, whatever, this is the future. It's really fun and interesting to watch the speed of development. I've made a few multi agent coding setups in the last", "context": "", "use_case": "", "switched_from": "", "story": "Welcome to Gas Town"}, {"id": 742871552, "item_id": 46429472, "url": "https://news.ycombinator.com/item?id=46429472", "by": "frankc", "date": "2025-12-29", "stance": "recommend", "quote": "I have been using beads for a few days on a couple of projects.", "excerpt": "Personally, I have been using beads for a few days on a couple of projects. I also like https://github.com/Dicklesworthstone/beads_viewer which is a nice tui for beads (with some additional workflow i haven't tried). I have found its been useful for longer, multi-session implementations. Its easier to get back into the work. I wouldn't go so far as to it couldn't do the work without it, but so far it seems smoother. ", "context": "", "use_case": "multi-session agent task tracking", "switched_from": "", "story": "Stop Claude Code from forgetting everything"}, {"id": 741412752, "item_id": 46338297, "url": "https://news.ycombinator.com/item?id=46338297", "by": "giancarlostoro", "date": "2025-12-20", "stance": "recommend", "quote": "I use Beads with Claude Code + Zed, and just brain dump to Claude to make a bunch of tasks with Beads", "excerpt": "I use Beads with Claude Code + Zed, and just brain dump to Claude to make a bunch of tasks with Beads, I have Claude ask me clarifying questions, then I let Claude go to town. Sometimes I let Claude work in parallel but APC seems to sometimes not like that not sure why.", "context": "", "use_case": "task tracking for agents", "switched_from": "", "story": "Show HN: Claude Code Plugin to play music when waiting on user input"}, {"id": 737293264, "item_id": 46080829, "url": "https://news.ycombinator.com/item?id=46080829", "by": "jauntywundrkind", "date": "2025-11-28", "stance": "recommend", "quote": "Somewhat aside but I love the data architrcture.", "excerpt": "Somewhat aside but I love the data architrcture. Jsonl lines checked into git, and a sqlite local cache. Auto-sync to replicate changes between the canonical and cache. https://github.com/steveyegge/beads?tab=readme-ov-file#the-magic-distributed-database-via-git I finally started digging in to OpenCode for real these past couple weeks. It has a planning mode, which nicely builds out a plan on text chat as usual, but ", "context": "", "use_case": "agent memory and task tracking", "switched_from": "", "story": "Beads \u2013 A memory upgrade for your coding agent"}, {"id": 737264529, "item_id": 46079033, "url": "https://news.ycombinator.com/item?id=46079033", "by": "hmokiguess", "date": "2025-11-28", "stance": "neutral", "quote": "My approach is to combine beads with superpowers skills", "excerpt": "First I hear of spec-kit, that looks very promising, I\u2019m interested in trying it. My approach is to combine beads with superpowers skills https://github.com/obra/superpowers I\u2019m wondering how does it compare to this, gonna give it a try, thanks!", "context": "", "use_case": "coding agent task tracking", "switched_from": "", "story": "Beads \u2013 A memory upgrade for your coding agent"}, {"id": 735011712, "item_id": 45938232, "url": "https://news.ycombinator.com/item?id=45938232", "by": "qudat", "date": "2025-11-15", "stance": "recommend", "quote": "Just use beads and you\u2019ll see what I mean.", "excerpt": "I\u2019ve done spec driven development using a bunch of markdown files. That works fine but what I have found really works is using beads: https://github.com/steveyegge/beads I\u2019m letting the agent help me draft the specs anyway and I found that the agent is a lot more focused when it can traverse a task tree using beads. It\u2019s the one spec or planning tool that I find really helps get things done without a bunch of interve", "context": "", "use_case": "agent task tracking and planning", "switched_from": "markdown spec files", "story": "Spec-Driven Development: The Waterfall Strikes Back"}]}