{"slug": "show-hn-lowfat-pluggable-cli-filter-that-saved-91-8-of-my-llm-tokens", "title": "Show HN: Lowfat – pluggable CLI filter that saved 91.8% of my LLM tokens", "summary": "A developer has released Lowfat, a pluggable CLI filter tool that reduced their LLM token usage by 91.8% over two months. The single-binary tool sits between CLI commands and AI agents, stripping verbose output like full YAML dumps to only pass relevant data, with a plugin system for customizing filters per command. The project aims to help users avoid token limits while maintaining local-first, telemetry-free data ownership.", "body_md": "Hi HN,\n\nNot sure if anyone would be interested.\n\nBut, just wanted to share that I've been maintaining my small tool called 'lowfat' that helps me filters some of my verbose CLI output.\n\nIt's a single binary, works as an agent hook or a shell wrapper. It has a plugin system to customize filters per command.\n\nThe idea is pretty simple: agents don't need the full kubectl get -o yaml or any 10k-line dump to make decisions. So that lowfat sits in between, strips the noise, and passes through what matters.\n\nHere's my real report after 2 months of personal use:\n\nlowfat history --all\n\n```\n  lowfat plugin candidates\n  ─────────────────────────────────────────────────────────\n\n    #  command                    runs   avg raw      cost   savings  source    status  \n    1  kubectl get                101x     14.4K      1.5M     93.9%  plugin    good    \n    2  grep                       103x     13.5K      1.4M     96.2%  plugin    good    \n    3  git diff                    81x       995     80.6K     57.9%  built-in  good    \n    4  kubectl                     90x       485     43.6K     33.6%  plugin    good    \n    5  docker                     127x      5.5K    693.6K     96.1%  built-in  good    \n    6  ls                         489x       117     57.3K     56.2%  built-in  good    \n    7  find                        30x     16.5K    495.0K     95.5%  plugin    good    \n    8  git show                    63x       490     30.9K     38.0%  built-in  good    \n    9  git                        177x       368     65.2K     76.1%  built-in  good    \n   10  git log                     86x       556     47.8K     78.5%  built-in  good    \n   11  kubectl logs                 5x      3.6K     17.8K     43.0%  plugin    good    \n   12  git status                  86x       152     13.1K     58.0%  built-in  good    \n   13  docker ps                   20x       467      9.3K     52.8%  plugin    good    \n   14  kubectl describe             6x       656      3.9K      1.2%  plugin    weak    \n   15  docker images                9x       940      8.5K     61.8%  built-in  good    \n   16  k get                        2x      2.1K      4.2K     35.9%  plugin    good    \n   17  terraform                   10x       395      3.9K     32.1%  plugin    good    \n   18  git commit                  32x        77      2.5K      0.0%  built-in  weak    \n   19  docker build                 8x       487      3.9K     37.6%  built-in  good    \n   20  docker compose              22x       979     21.5K     89.4%  built-in  good    \n\n  total: 4.4M raw → 4.1M saved (91.8%)\n```\n\nMy toolset above is kind limited, but it works pretty well for my usecase without any interruption Kinda help me not reaching the token limit for my company Bedrock limit usage and keep optimizing the saving on the go for later usage.But, why not alternatives ([https://github.com/zdk/lowfat#alternatives](https://github.com/zdk/lowfat#alternatives)) ?\nThe answers are:\n- My goal is to make the core lightweight but extensible via plugins i.e. not trying to bundle every command in the installed binary so that people own their output filters.\n- Customizable per usecase via plugin or filter pipelines as I am using my own toolset.\n- Customizable for non-public CLI tools, for example, some enterprise might have their interal CLI tools that public won't have access.\n- People should own their data. So the design is local-first, No telemetry forever.\n- I kinda love UNIX-style composible pipes, so lowfat-filter has implemented this style.\n- Be able to adjust aggressiveness of the filter, so we can control that we won't strip something the agent needed.\n\nGitHub: [https://github.com/zdk/lowfat](https://github.com/zdk/lowfat)\n\nAnyway, if anyone is interested, feedbacks and questions are welcome!\n\nThanks!\n\nComments URL: [https://news.ycombinator.com/item?id=48409955](https://news.ycombinator.com/item?id=48409955)\n\nPoints: 1\n\n# Comments: 0", "url": "https://wpnews.pro/news/show-hn-lowfat-pluggable-cli-filter-that-saved-91-8-of-my-llm-tokens", "canonical_source": "https://github.com/zdk/lowfat", "published_at": "2026-06-05 09:10:42+00:00", "updated_at": "2026-06-05 09:51:34.022292+00:00", "lang": "en", "topics": ["ai-tools", "large-language-models", "ai-agents"], "entities": ["lowfat", "kubectl", "docker", "git", "grep", "find", "LLM"], "alternates": {"html": "https://wpnews.pro/news/show-hn-lowfat-pluggable-cli-filter-that-saved-91-8-of-my-llm-tokens", "markdown": "https://wpnews.pro/news/show-hn-lowfat-pluggable-cli-filter-that-saved-91-8-of-my-llm-tokens.md", "text": "https://wpnews.pro/news/show-hn-lowfat-pluggable-cli-filter-that-saved-91-8-of-my-llm-tokens.txt", "jsonld": "https://wpnews.pro/news/show-hn-lowfat-pluggable-cli-filter-that-saved-91-8-of-my-llm-tokens.jsonld"}}