Topic
#go
Loot, blog posts and adjacent themes connected to this topic. Follow the tag to keep it in your orbit.
4Shown loot
0Shown articles
8Linked neighbor tags
Topic paths
If you want to go deeper, the adjacent tags are the fastest way to compare and branch into related workflows.
Loot
More from this topic
Text: AI-generated
AI-generated · Automatically published by LinkLoot. A Go server that speaks the real Xet protocol, so the standard hf CLI works against your own endpoint — with a pull-through proxy that keeps cached repos available even when huggingface.co goes down. AI-generated: This Loot was created and published automatically by LinkLoot and was not substantively reviewed by a human editor. What it does xet-server is a Go implementation of Hugging Face's Xet storage protocol, the content-defined-chunking, dedup-first layer that replaced Git LFS for large model files on the Hub. It is wire-compatible: point the real hf CLI (huggingfacehub + hfxet) at it via HFENDPOINT, and hf upload / hf download work end-to-end, producing byte-identical round-trips with genuine BLAKE3 hashing, real xorb/shard binary formats, and LZ4/byte-grouping compression. The project started as a simplified chunking demo and grew into a from-scratch, spec-driven reimplementation, verified against a live hfxet client. Several wire-format quirks were found only by capturing and replaying real client traffic (documented in docs/PROTOCOL.md). It ships two main pieces: The wire-compatible server — a CAS HTTP API (xorb upload/fetch, shard upload, file reconstruction with Range paging) plus a Hub API shim covering repo create, preupload, read/write tokens, commit, and resolve/HEAD, including independent revisions and branches per repo. xet-proxyd — a caching pull-through proxy for the real huggingface.co. This is an offline-resilience layer, not just a performance cache: any repo metadata or xorb bytes it has relayed once stay servable even after the upstream becomes unreachable, and its cache directory can be handed to a plain xetd as a permanent disconnected replacement. Storage backends are pluggable, and interactive Swagger UI docs are included. Who it helps Teams running air-gapped or bandwidth-constrained ML infrastructure who want to serve model weights through the standard toolchain without depending on Hugging Face's hosted endpoint. Also useful for anyone needing a local staging target for Xet-based workflows, or an on-prem resilience layer that survives upstream outages. Getting started Build and run the server, then set HFENDPOINT to its address and use the normal hf commands. The README walks through running the wire-compatible server, pointing the CLI at it, the Xet Data API + CLI, and xet-proxyd. A troubleshooting section covers common issues, and a "Where this diverges from real Xet" section documents known gaps. Limits and costs MIT-licensed, so source use is unrestricted, but hosting is on you — hardware, storage, and operational overhead are not covered by the license. It is a from-scratch reimplementation, so expect documented divergences from the production Xet service; review those before relying on it in critical paths. No pricing is published because there is none beyond self-hosting. Source links Official repository: https://github.com/guilt/xet-server Project documentation: https://github.com/guilt/xet-server/blob/main/README.md Project license: https://github.com/guilt/xet-server/blob/main/LICENSE.md
- LinkLoot access
- Free
- Provider costs
- Unknown
Review open
Text: AI-generated
AI-generated · Automatically published by LinkLoot. Gonic is a free-software Go implementation of the Subsonic API that streams your own collection to dozens of compatible clients, runs on modest hardware like a Raspberry Pi, and adds podcasts, multi-user support, scrobbling, and gapless jukebox playback. > AI-generated: This Loot was created and published automatically by LinkLoot and was not substantively reviewed by a human editor. What it does Gonic is a self-hosted music streaming server that implements the open Subsonic server API. Because it speaks that protocol, it works with the wide ecosystem of Subsonic-compatible clients rather than locking you into one app. It indexes your existing folder tree and preserves it for browsing, while also letting you navigate by tags through taglib — covering mp3, opus, flac, ape, m4a, wav and more. Audio can be transcoded on the fly via ffmpeg and cached, and a server-side jukebox mode enables gapless playback instead of streaming. Podcasts are supported alongside music. Other documented features include multiple users each with their own playlists, top tracks and transcoding preferences; last.fm and ListenBrainz scrobbling; artist similarities and biographies pulled from the last.fm API; handling of multi-valued tags such as albumartists and genres; and an album-artist tag so compilation appearances don't clutter your artist list. A web interface handles configuration: setting up integrations, managing users and starting scans. The project reports fast scanning — roughly 10 minutes for an initial pass over about 50k tracks and around 6 seconds for incremental rescans — though real-world results depend on your library and hardware. Who it helps Anyone who owns their music files and wants remote access without renting space or handing a third party their collection. Written in Go, Gonic is described as lightweight and suitable for single-board computers like a Raspberry Pi, with ARM images available. Households benefit from per-user accounts; listeners who use apps such as Ultrasonic, DSub, Symfonium, Amperfy, Jamstash, Soundwaves or Strawberry get native client support. Getting started Point it at your collection with GONICMUSICPATH (or -music-path). Optional paths cover podcasts (GONICPODCASTPATH), playlists (GONICPLAYLISTSPATH, using numbered /.m3u layout) and a transcode cache (GONICCACHEPATH). The default login is admin/admin; change the password from the web interface once you're in. Community discussion happens on the gonic Libera.Chat channel. Limits and costs The software is GPL-3.0 licensed, but that refers to the code's freedom, not hosting: running your own server still costs electricity, storage and any bandwidth you pay for. Transcoding requires ffmpeg installed on the host. Client compatibility is based on projects the maintainers tested against, so behavior in other apps may vary. Pricing beyond a self-managed install is not specified. Source links Official repository: https://github.com/sentriz/gonic Project documentation: https://github.com/sentriz/gonic/blob/master/README.md Project license: https://github.com/sentriz/gonic/blob/master/LICENSE
- LinkLoot access
- Free
- Provider costs
- Unknown
Review open
Give your editor an agent that streams, asks before it touches files, and survives restarts
No votes yet
Text: AI-generated
AI-generated · Automatically published by LinkLoot. acplause is a Go-based Agent Client Protocol v2 agent that connects any ACP-compatible editor to OpenRouter or another OpenAI-compatible API — with permission gates, session persistence, and live terminal output. AI-generated: This Loot was created and published automatically by LinkLoot and was not substantively reviewed by a human editor. What it does acplause is an experimental agent built on the Agent Client Protocol (ACP) v2. Your editor talks to it over stdio; it thinks using any OpenAI-compatible Chat Completions API — OpenRouter by default — and acts inside your workspace through tools. It streams text and reasoning tokens back to the editor, surfaces tool calls, usage, and cost reporting in real time, and shows live bash terminal output with bounded snapshots for session replay. Before changing a file or running a command, it asks permission. Sessions survive process restarts and can move between editor windows. It supports model and reasoning-effort selection from OpenRouter's model list, PNG/JPEG prompt attachments, MCP tools over stdio and Streamable HTTP, structured user questions via ACP form elicitation, native execution plans with persistent progress, per-session Plan/Build modes with enforced tool restrictions, and a /compact command to summarize older context on demand. Who it helps Developers who want an AI coding agent wired into their existing editor rather than a separate chat window. If you already use an ACP v2 client, acplause slots in as the backend without replacing your workflow. The permission-before-write model suits anyone cautious about letting an agent modify files unsupervised. Cost and usage reporting helps teams tracking token spend across sessions. Getting started Three install paths exist: Homebrew (brew install anesthetised/tap/acplause) builds the tagged source locally, installing Go as a build dependency. On macOS this avoids requiring a paid Developer ID certificate, at the cost of a slower first install including SQLite compilation. Prebuilt binaries for macOS and Linux (arm64/amd64) are available from GitHub Releases with SHA256 checksums and artifact attestations. Note: prebuilt macOS binaries are not Developer ID signed or notarized, so macOS may block them until you explicitly allow execution. Go install (go install github.com/anesthetised/acplause/cmd/acplause@latest) if you prefer the standard toolchain. Configure your ACP v2 client to launch acplause with ACPLAUSEAPIKEY set to your provider key. Set ACPLAUSEBASEURL for a non-OpenRouter endpoint. Without a key, no Authorization header is sent, which works with local providers that don't require auth. Run acplause doctor to check local configuration and data directories, or acplause doctor --online to also verify provider connectivity. Limits and costs acplause is explicitly experimental. Calendar-versioned releases do not promise compatibility; breaking changes appear in release notes. It follows the pinned ACP v2 schema only. No pricing information is provided in the documentation — model costs depend entirely on whichever API provider you configure. The MIT license covers the software itself but says nothing about hosting or API usage fees. Source links Official repository Project documentation (README) MIT License
- LinkLoot access
- Free
- Provider costs
- Unknown
Review open
PicoClaw is a fascinating ultra-light agent project — but it is not a clean 1:1 OpenClaw replacement
No votes yet
PicoClaw offers a lightweight AI agent experience built for diverse hardware, emphasizing compact design and broad architecture support. The project highlights fast startup and flexible deployment options, making it appealing for developers targeting low-cost systems. Yes — this is worth a Loot, because the hardware and footprint story is genuinely interesting. PicoClaw makes a credible case for an ultra-light AI agent stack in Go that can run on extremely cheap hardware, with fast startup and wide architecture support. What looks genuinely strong pure Go implementation very broad platform story: RISC-V, ARM, MIPS, x86, Android claimed <10MB core footprint in early builds, though the repo also says recent builds can hit 10–20MB local launcher, Docker path, Telegram/gateway flow, and multi-provider support ambitious feature surface for such a small runtime The critical reality check The viral framing overshoots the evidence. The repo itself says: early rapid development do not deploy to production before v1.0 unresolved security issues may still exist memory usage has already drifted upward in recent builds So the real story is promising lightweight agent engineering, not a fully proven OpenClaw killer.
- LinkLoot access
- Free
- Provider costs
- Unknown
Review open
Blog
Related reads
No blog posts for #go yet
There is no published article with this tag right now. Browse the blog for adjacent themes or follow the tag for future updates.