惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

AI
AI
Cloudbric
Cloudbric
Schneier on Security
Schneier on Security
V2EX - 技术
V2EX - 技术
N
News and Events Feed by Topic
Hacker News: Ask HN
Hacker News: Ask HN
L
LINUX DO - 最新话题
V
Visual Studio Blog
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
W
WeLiveSecurity
博客园 - 司徒正美
Jina AI
Jina AI
S
Secure Thoughts
罗磊的独立博客
Hugging Face - Blog
Hugging Face - Blog
有赞技术团队
有赞技术团队
WordPress大学
WordPress大学
Security Archives - TechRepublic
Security Archives - TechRepublic
博客园 - 三生石上(FineUI控件)
The Last Watchdog
The Last Watchdog
S
Security @ Cisco Blogs
S
SegmentFault 最新的问题
Attack and Defense Labs
Attack and Defense Labs
Exploit-DB.com RSS Feed
Exploit-DB.com RSS Feed
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
SecWiki News
SecWiki News
Google DeepMind News
Google DeepMind News
T
Troy Hunt's Blog
H
Heimdal Security Blog
C
CXSECURITY Database RSS Feed - CXSecurity.com
N
News and Events Feed by Topic
雷峰网
雷峰网
Last Week in AI
Last Week in AI
IT之家
IT之家
Project Zero
Project Zero
O
OpenAI News
腾讯CDC
The Hacker News
The Hacker News
L
Lohrmann on Cybersecurity
T
Tenable Blog
博客园_首页
C
Cisco Blogs
酷 壳 – CoolShell
酷 壳 – CoolShell
S
Schneier on Security
Scott Helme
Scott Helme
cs.CL updates on arXiv.org
cs.CL updates on arXiv.org
人人都是产品经理
人人都是产品经理
P
Privacy International News Feed
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻

Show HN

GitHub - flightdeckhq/flightdeck: Observability and control plane for AI agents. CSP Radar GitHub - Light-Heart-Labs/DreamServer: Turn your PC, Mac, or Linux box into an AI server. LLM inference, chat UI, voice, agents, workflows, RAG, and image generation. GitHub - Diplomat-ai/diplomat-agent-ts: What can your TypeScript AI agent do to the real world? Scan your code. See which tool calls have zero checks Code Block Selector - Visual Studio Marketplace Prometheus dependency graph — interactive showcase | Riftmap Show HN: I made a vi-like modal keyboard plugin for Figma GitHub - run-llama/liteparse: A fast, helpful, and open-source document parser GitHub - dalemyers/Roar: A macOS CLI tool for notifications GitHub - district-solutions/open-agent-tools-coder: Enables small-to-large self-hosted ai models to use local source code when running tool-calling agentic workloads. We actively data mine 20,900+ (2+ TB) popular github repos using large and small ai models to create reuseable: json, markdown and parquet files for local-first tool-calling models. GitHub - progapandist/stripeek: A local TUI proxy for real-time Stripe API debugging, built for navigating complex payloads fast. GitHub - sir1st/hermes-desktop: All-in-one cross-platform desktop app for Hermes Agent — bundles Python + hermes-agent + hermes-web-ui GitHub - astefanutti/shaderbang: Shebang for Shaders Show HN: Generate Claude Code Workflows using Spec Driven Development approach GitHub - nixys/nxs-universal-chart: The Helm chart you can use to install any of your applications into Kubernetes/OpenShift Show HN: AI agents for UK GDAD PCF roles and their skills The Two Pillars: Mixer Mode and Meta-Software in the Reorganization of Software Work After AI GitHub - JaiCode08/teleport-env What 1,000+ Harness Experiments Taught Me About Self-Improving Agents Show HN: Liiists, a Markdown-first, iOS and CLI list app SwiperTab – Get this Extension for 🦊 Firefox (en-US) GitHub - kouhxp/fftext: Summarize, explain, fact-check, or translate any text, URL, or file. No GPU. No cloud. One command GitHub - sweetpad-dev/sweetpad: Develop Swift/iOS projects using VSCode GitHub - dogmaticdev/IRON: IRON a.k.a. Intermediate Representation Object Notation is a Interpreter/Database that is used to create Programming Languages. GitHub - sjhalani7/vaen: Package your AI coding harness into a portable .agent file, and share it across repos, teams, & the community without ever having to copy-paste instructions, skills, MCP config, or secrets. Show HN: Gandalf the Grader Show HN: Citadeld – replay any CI failure locally from a single file GitHub - tdortman/cuSBF: High-Performance GPU Super Bloom Filter coral-ai/claude-code-token-xray at main · Coral-Bricks-AI/coral-ai GitHub - ulyssestenn/funes: Funes is a Git-based framework for LLM-managed knowledge work: an AI Librarian ingests raw sources, builds an interlinked Markdown knowledge base, and uses it to produce cited reports, analyses, and other outputs. GitHub - ThatXliner/gah: Git Add Hunk, built for agents to use GitHub - harmont-dev/harmont-cli: Command-line client for the Harmont CI platform GitHub - brooksmcmillin/mcp-authflow: OAuth 2.0 Authorization Server framework for MCP servers GitHub - javaid-codes/audit-supply-chain-agents GitHub - amorey/gochan: A small library of common channel architectures for Go, inspired by Rust GitHub - arifozgun/OpenGem: Free, Open-Source AI API Gateway with Gemini, OpenAI & Anthropic Compatibility in 1 file GitHub - Pranesh950/BioPetals: 🌸 Run BIOxAI models at home, BitTorrent-style. Fine-tuning and inference up to 10x faster than offloading GitHub - cnguyen14/bounty-doctor: Diagnose a GitHub bounty issue before you waste hours: detects honeypot scam repos, AI-bot attempt swarms, and stale contests. Show HN: CoreMCP – MCP Server for On-Prem DBs Show HN: KittyHTML – Render HTML/CSS as an inline image in your terminal GitHub - bingud/filemat: Web-based file manager Show HN: TruthLens – Free multi-signal deepfake image detector GitHub - apexlocal-jz/claude-usage-tray: Windows system-tray app showing your Claude Code rate-limit usage at a glance. Zero deps, ~300 lines of PowerShell. Cross-IDE (works regardless of VS Code, Cursor, plain terminal). Release v0.1.2.1 · kouhxp/yapsnap GitHub - noopolis/moltnet: Self-hostable chat network for AI agents. Pre-built bridges for Claude Code, Codex, and the Claws. Rooms, DMs, history. No Slack bots, no Matrix, no glue code. GitHub - tamerh/enju: Coordinating Humans, AI Agents, and Compute as Peers on a Shared Workflow Graph Show HN: Continuity-auth – Respect-weighted rate limits for the open web GitHub - luml-ai/luml: AI lifecycle platform where engineers and agents track experiments, train models, and ship to production. GitHub - mrdanielcasper/CoreTex: A UNIX-inspired, biomimetic, flat-file AI harness and knowledge engine. GitHub - clemg/pierre-github: Pierre's diffs.com and trees.software for Github GitHub - lyriks-io/unspaghettit: Behavior-driven AI development without prompt spaghetti. GitHub - sofumel/claude-handoff-revive: Resume Claude Code work after rate/usage/context limits without replaying the prior transcript. Auto-saves at 90%/95% usage. Plugin-installable, 10 languages. GitHub - dotexorg/saferpc: Typed, end-to-end encrypted RPC over any bidirectional channel. GitHub - BeeZeeAgent/beezee: Agent harness orchestration Legato Next.js Boilerplate for Internal Tools · CoreUI GitHub - clark-labs-inc/clark-hash: Clark Hash, 32x smaller searchable sketches for embeddings GitHub - ZeroPointRepo/youtube-mcp: The fastest YouTube transcript + YouTube search MCP for AI agents. Try for free. Typing Mastery — climb toward 100+ WPM, deliberately GitHub - Andebugulin/Awareen GitHub - fayzan123/claude-workflow-composer: Visual desktop app for composing multi-agent coding workflows. Drag agents, attach skills and MCPs, wire handoffs, export to .claude/ GitHub - harshaneel/humanize: Best static AI text humanizer. Two research-grounded skills that work in any LLM (Claude, ChatGPT, Gemini, Codex): humanize beats perplexity-based detectors, ai-check produces forensic scoring with evidence-quoted flags. Nine levers, 50+ peer-reviewed sources, 2024-2026 detection literature. GitHub - StackOneHQ/stack-nudge GitHub - nodes-app/swift-markdown-engine: A native AppKit Markdown editor for macOS, built on TextKit 2 and bridged to SwiftUI. We hardened an LLM agent. Each defense we added made it more exploitable. GitHub - alkait/WhatsKept: Agent-queryable WhatsApp history from an iOS backup — a single Go binary. GitHub - octelium/cordium: Open-source, general-purpose sandbox platform for devs and AI agents that provides identity-based secure access to infrastructure without credentials. WAR.GOV/UFO Microfilm5 GitHub - scosman/videowright: Build animated explainer videos with your coding agent GitHub - dipankar/dscode: The code editor you can take apart. GitHub - zoharbabin/web-researcher-mcp: MCP server (Go) for AI assistants: web search, content extraction, academic/patent/news research. Multi-provider routing, 4-tier scraping, search lenses. Works with Claude, Cursor, and any MCP client. GitHub - ruvnet/RuView: π RuView turns commodity WiFi signals into real-time spatial intelligence, vital sign monitoring, and presence detection — all without a single pixel of video. GitHub - scanaislop/aislop: Catch the slop AI coding agents leave in your code: narrative comments, swallowed exceptions, as-any casts, dead code, oversized functions. 50+ rules across 7 languages (TypeScript, JavaScript, Python, Go, Rust, Ruby, PHP). Sub-second, deterministic, no LLM at runtime. MIT-licensed. GitHub - kouhxp/cheap-im: CPU-only voice agent approximating Thinking Machines' Interaction Models demo GitHub - unprovable/OrchidMantis: Orchid Mantis — standalone framework for Zero-Knowledge Proofs of eXploit (ZKPoX). GitHub - MarcellM01/TinySearch: Shrink the web for your local LLMs! GitHub - pileax-ai/pileax: PileaX is an all-in-one AI knowledge base system. 🍀 GitHub - TangibleResearch/Halgorithem: A Algo designed to detect AI Hallucitions GitHub - DO-SAY-GO/freelang: I love freelang GitHub - CarpseDeam/Aura-IDE: An AI coding harness that shaped itself - Planner/Worker agents, repo awareness, surgical edits, validation, recovery, and safe diff approvals. GitHub - chojs23/concord: A feature-rich TUI client for Discord GitHub - tommyjepsen/awesome-ux-skills: UX & AI Product designs skills you can use today in Claude Code GitHub - aerf-spec/aerf: Agent Evidence Receipt Format (AERF) — an open specification for tamper-evident, independently verifiable records of AI agent actions. GitHub - kklimuk/docx-cli: CLI for AI agents (Claude, Codex) to read, edit, and comment on .docx files with full format fidelity. GitHub - Jwrede/tokentoll: Catch LLM cost changes in code review. Infracost for LLM spend. GitHub - samchon/ttsc: A `typescript-go` toolchain for compiler-powered plugins and type-safe execution + 500x faster lint integrated into compiler GitHub - Higangssh/homebutler: 🏠 Manage your homelab from chat. Single binary, zero dependencies. GitHub - olalie/tapmap: See where your computer connects and what stands out on a live world map. GitHub - matisiekpl/neond: DX-focused control plane for Postgres dedicated to non-critical workloads. Your postgres:latest replacement 🐘 GitHub - Diplomat-ai/diplomat-agent: What can your AI agent do to the real world? Scan your code. See which tool calls have zero checks GitHub - Bajusz15/beacon: Open-source agent for secure remote access, monitoring, and deploys across home-lab and self-hosted machines like Raspberry Pi, N100, or any Linux server. Open web based TTY or tunnel Home Assistant and other local services securely without opening ports. BigTech AI News - Chrome 应用商店 GitHub - vinhnx/VTCode: VT Code is an open-source coding agent with LLM-native code understanding and robust shell safety. Supports multiple LLM providers with automatic failover and efficient context management. GitHub - michaelaz774/decision-engine: A decision operating system for startup founders, powered by Claude Code. Synthesizes wisdom from 25+ legendary founders and investors into interactive AI-driven decision frameworks. GitHub - Chrilleweb/dotenv-diff: Validate environment variable usage in your codebase GitHub - Lumen-Labs/brainapi2: BrainAPI is a knowledge graph–powered AI memory layer that transforms unstructured data into structured knowledge, enabling intelligent search, recommendations, and contextual memory for AI agents and applications. GitHub - familiar-software/familiar: Let AI watch you work. Familiar lets your AI update its memory, skills, and knowledge by watching your screen. GitHub - skorotkiewicz/rudo: A small, elegant dock for Wayland GitHub - muxshed/shed: One stream in, or many. Every destination, simultaneously. No cloud middleman, no per-channel fees, no limits. make sidebar/address bar rounded corner toggleable
GitHub - momiji-rs/sasso: A pure-Rust SCSS to CSS compiler (a dart-sass alternative). Zero dependencies, wasm-friendly, lib + CLI.
linyiru · 2026-06-28 · via Show HN

crates.io docs.rs CI CodSpeed sass-spec dart-sass runtime deps license

A pure-Rust SCSS → CSS compiler — a from-scratch dart-sass alternative. Zero runtime dependencies, wasm-friendly, usable as a library and a CLI, and designed to match current dart-sass byte-for-byte on the subset it implements.

Status: v0.x, maturing fast. Compiles real-world SCSS and indented .sass byte-identically to dart-sass 1.101, and passes 100% of the attempted official sass-spec suite (13,896 / 13,896, zero failures) — tracked as a ratchet (see Conformance for exactly what that denominator means). What remains is real-world breadth and hardening, not spec coverage.

Why another Sass compiler?

grass is the incumbent Rust implementation and a strong one (it compiles Bootstrap/Bulma byte-accurately and is ~2× faster than dart-sass). But it is pinned to dart-sass 1.54.3 (mid-2022) and predates the CSS Color Level 4 overhaul, so it diverges from current dart-sass on, e.g., fractional color channels (rgb(63.75, 127.5, 191.25) vs rounded hex) and emits hex where dart-sass now keeps rgb()/hsl() forms. sasso targets current dart-sass exactly, with a span-first parser, a modern color model, and a zero-dependency, sandbox-friendly core. See docs/GRASS_LANDSCAPE.md for the full analysis.

Features (this slice)

  • $variables, lexical scoping, !default, !global
  • Nesting, the & parent selector (with selector-list multiplication), and combinator normalization (>, +, ~)
  • #{} interpolation in selectors, property names and values
  • // (stripped) and /* */ (preserved) comments
  • Numbers with units and unit arithmetic ($pad * 2 → 16px)
  • A full color model with fractional channels + author-spelling preservation (red, #336699, rgb()/hsl() round-trip unchanged)
  • Color functions: rgb/rgba/hsl/hsla/mix/lighten/darken/ percentage (+ red/green/blue/alpha)
  • @import partial inlining through a pluggable Importer (CSS imports pass through)
  • expanded and compressed output styles
  • Source maps (v3) — byte-exact to dart-sass 1.101, on the library (compile_with_source_map), the CLI (--source-map), and wasm (compile(scss, { sourceMap: true }))
  • Verbatim preservation of CSS functions it doesn't own (calc, var, clamp, translateX, …)

Since this slice was written the ratchet has added a great deal more — @mixin/@function, control flow (@if/@each/@for/@while), @extend and %placeholders, a calc() engine, the CSS unit system + math functions, full CSS Color 4 color spaces (oklch/lab/color()…), structured @media/@supports, maps, the @use/@forward module system (built-in sass:* modules + user files), and the indented .sass syntax. The compiler now passes 100% of the attempted sass-spec suite (13,896 / 13,896, zero failures) byte-for-byte against dart-sass 1.101 — 11,405 byte-exact CSS outputs plus 2,491 error specs it correctly rejects (see Conformance).

Install

CLI — prebuilt binaries. Every release ships static binaries for Linux (gnu + musl), macOS and Windows (x86_64 / aarch64), built with cargo-dist. Grab one from the Releases page, or:

$ curl -fsSL https://github.com/momiji-rs/sasso/releases/latest/download/sasso-installer.sh | sh   # Linux/macOS
$ cargo binstall sasso        # fetch the prebuilt binary
$ cargo install sasso         # build from source (needs a Rust toolchain)

Library — crates.io.

WebAssembly — npm. A tiny, dependency-free wasm build that mirrors the dart-sass modern JS API, so it's a drop-in for the sass npm package in build tools (no wasm-bindgen, no native add-ons):

import { compileString } from "sasso";
compileString("a { color: #ffffff }", { style: "compressed" }).css; // a{color:#fff}

Works as the sass implementation in webpack/sass-loader (implementation: require("sasso"), api: "modern") and Vite (alias "sass": "npm:sasso").

Library usage

use sasso::{compile, Options, OutputStyle};

let scss = "$c: #336699; .a { color: $c; &:hover { color: lighten($c, 10%); } }";
let css = compile(scss, &Options::default()).unwrap();
assert!(css.contains("a:hover"));

// Minified:
let min = compile(scss, &Options::default().with_style(OutputStyle::Compressed)).unwrap();

@import resolution is controlled by an Importer you supply, so file access stays on your side of any sandbox:

use sasso::{compile, Importer, Options};

struct MyFs;
impl Importer for MyFs {
    fn resolve(&self, path: &str) -> Option<String> {
        std::fs::read_to_string(format!("scss/_{path}.scss")).ok()
    }
}
let css = compile("@import \"base\";", &Options::default().with_importer(&MyFs)).unwrap();

A ready-made FsImporter is provided for standalone/CLI use.

CLI usage

$ cargo install --path .            # installs the `sasso` binary
$ sasso input.scss              # CSS to stdout (expanded)
$ sasso --style=compressed input.scss
$ sasso -I scss/ main.scss      # add @import load paths
$ echo '.a{color:red}' | sasso --stdin

Conformance

The official sass-spec suite is the parity oracle. The harness in spec/ runs the compiler against every spec case and reports a pass rate; we ratchet it upward over time.

Metric Value
sass-spec commit 1b03109a (dart-sass 1.101.0)
Total cases 13,904
Attempted (excl. 8 dart-sass :todo) 13,896
Passing 13,896 — 100% of attempted · 0 failures (99.94% of all 13,904)
↳ byte-exact CSS output 11,405
↳ error specs correctly rejected 2,491

Passing = byte-exact CSS output match plus error specs the compiler correctly rejects — the standard sass-spec conformance metric (the harness checks that an error spec errors; the error message is tracked separately as a non-gating metric). The 8 excluded cases are tagged :todo for dart-sass itself upstream — dart-sass doesn't pass them either; sasso matches dart-sass's actual behaviour on all 8 regardless.

Strict input validation, too. Matching dart-sass means rejecting what dart-sass rejects, not just reproducing its output. sasso errors — rather than silently accepting — on an invalid hex literal (#00000), out-of-grammar rgb()/hsl() arguments, a duplicate @mixin/@function parameter, a misplaced @content/@extend, a style rule or declaration inside a @function body, a malformed :nth-child() / empty :not() selector, a bad @charset or @at-root (…) query, and more — each with dart-sass's exact message.

Run it yourself:

$ spec/fetch.sh                                      # clone the suite
$ cargo build --release
$ SASS_BIN=target/release/sasso python3 spec/run_spec.py

Performance

sasso is a native, in-process library — no subprocess, no Node, no Dart VM — so startup is effectively free, which dominates when a build compiles many files. On an Apple M2 Max it is the fastest of the three engines measured, beating dart-sass by 19–30× end-to-end and leading grass (the incumbent Rust compiler) by ~2.3–2.9×:

Axis sasso grass dart-sass (bin) npx sass
Startup² 1.8 ms 1.8 ms 139 ms 495 ms
Cold single large file 11.7 ms 26.5 ms 354 ms 710 ms
Batch (40 files, 1 process) 49.0 ms 136 ms 916 ms
Pure compile (startup removed) 7.4 ms 21.3 ms ~216 ms¹

¹ derived (cold − startup) — dart-sass has no in-process loop mode. So sasso is ~29× faster than dart-sass on pure compute, ~30× on a cold single file, and ~77× on startup; vs grass it is ~2.3× cold / ~2.8× batch / ~2.9× pure.

² Startup compiles a 1-rule file, so it sits at the OS process-spawn floor — sasso and grass measure identically (1.8 ms) here (the mean is dominated by scheduler jitter at this sub-2 ms scale). The native library and wasm builds remove process startup entirely. A scoped bump-arena allocator (one audited unsafe module, Miri- and AddressSanitizer-verified; the rest of the library stays unsafe-free) gives a further ~1.5× by turning each compile's allocations into a pointer bump freed wholesale at the end. Composite values (strings, lists, maps) are reference-counted, so reading a $variable is an O(1) refcount bump, not a deep copy — on a large list passed through a call chain without mutation this cuts both instructions (~7×) and peak memory (~13×). A round of evaluator allocation trimming — skipping the per-rule selector clone when nothing extends it, iterating @each over the list's shared handle, and dropping redundant per-declaration copies — shaves a further ~2.8% off pure compile on representative stylesheets (measured by instructions-retired, since the win is below wall-clock jitter at this ms scale; byte-identical output). Full methodology, per-file numbers and the correctness diff are in bench/three_way.md; run it yourself with cd bench && RUNS=12 WARMUP=3 LOOP_N=200 bash scripts/run_bench.sh.

WebAssembly

Because the library is zero-dependency and pure std, it compiles to wasm32-unknown-unknown and wasm32-wasip1 out of the box (built in CI). The deployable .wasm cdylib ships in two variants, published to npm as sasso:

Variant Build Over the wire Compile (large, in Node)³
size (default) opt-level = "z" + LTO + panic = "abort" + strip + wasm-opt -Oz ~854 KB / ~356 KB gzip ~27 ms
speed (sasso/speed) opt-level = 3 + wasm-opt -O3 ~1.84 MB / ~637 KB gzip ~12 ms

³ in-process compile of the same large file, Node 22 (best-of-N). The wasm tax over native sasso (7.7 ms) is ~1.5× for the speed build and ~3.5× for the size build; the wasm build runs without the bump arena. Even so the speed build (~12 ms) beats native grass (21 ms) and every dart-sass form a Node toolchain can run.

A whole modern Sass compiler — @use/@forward, @extend, the calc engine, CSS Color 4 — in a few hundred KB gzipped, far smaller than shipping the dart-sass compiler as JavaScript. A browser playground is tracked in the issues.

Language bindings

sasso ships as a Rust crate and is usable from other languages too. First-party packages are released by this project and pin a published sasso crate version; community bindings are maintained in their own repos.

Language Package Maintained by How
Rust sasso (crates.io) First-party the core library — see Library usage above
JavaScript / wasm sasso (npm) First-party the in-repo wasm/ cdylib — see WebAssembly above
Ruby sasso (RubyGems) First-party momiji-rs/sasso-ruby — an in-process native extension (magnus + rb-sys) around this crate
Python sasso (PyPI) First-party momiji-rs/sasso-pythonctypes over the C ABI; one prebuilt wheel per platform, no build step
Go sasso-go (go get) First-party momiji-rs/sasso-gopure Go, no cgo: embeds the wasm build and runs it with wazero, so CGO_ENABLED=0 and cross-compilation just work. String-in/CSS-out (for file-based importers use the C ABI via cgo, below)
PHP shyim/php-sasso (PIE / pecl) Community (@shyim) an ext-php-rs extension (Sasso\Compiler) wrapping this crate in-process; prebuilt for PHP 8.2–8.5 (Linux glibc/musl, macOS)

Ruby framework integrations build on that gem — drop-in Sass for your stack, compiled in-process (no Node, no Dart, no subprocess) and byte-for-byte identical to dart-sass:

Framework Gem Repo
Rails (Propshaft + Sprockets) sasso-rails momiji-rs/sasso-rails
Bridgetown bridgetown-sasso momiji-rs/bridgetown-sasso
Hanami (2.1+) hanami-sasso momiji-rs/hanami-sasso

Each compiles Sass without a Node toolchain — typically ~6–7× faster per compile than the Node sass default (and far faster cold, with no process spawn).

C ABI — use sasso from any language

Beyond the packages above, sasso ships a C ABI (ffi/) so any language with a C FFI can drive the compiler in-process. Each release attaches a per-target sasso-<version>-<target>-c-api.tar.xz (.zip on Windows) containing the prebuilt library and the header — the universal substrate every binding sits on:

include/sasso.h
lib/  libsasso.a            # static — link it for a self-contained binary, no runtime dep
      libsasso.so|.dylib    # dynamic (Windows: sasso.dll + sasso.dll.lib + sasso.lib)

The ABI is two owned calls plus an optional importer callback — see ffi/include/sasso.h, the contract notes there, and runnable bindings for 8 languages under ffi/examples/ (C, Go, Ruby, Swift, Deno, Bun, LuaJIT, C#).

From Go

Statically link libsasso.a via cgo for a single self-contained binary (no CGO_ENABLED=0, but no runtime dependency either):

package main

/*
#cgo CFLAGS: -I./sasso-c-api/include
#cgo LDFLAGS: ./sasso-c-api/lib/libsasso.a
#include <stdlib.h>
#include "sasso.h"
*/
import "C"
import (
	"fmt"
	"unsafe"
)

func main() {
	src := ".a { .b { color: #336699 } }"
	cs := C.CString(src)
	defer C.free(unsafe.Pointer(cs))
	r := C.sasso_compile(cs, C.size_t(len(src)), nil)
	defer C.sasso_result_free(r)
	if r.ok == 0 {
		panic(C.GoString(r.error))
	}
	fmt.Print(C.GoStringN(r.css, C.int(r.css_len)))
}

A fuller cgo binding — options, errors, and a custom importer — is in ffi/examples/go/. Prefer no cgo? The first-party sasso-go package is ready-made: go get it and it embeds the wasm build and runs it with wazero, keeping CGO_ENABLED=0 and trivial cross-compilation (string-in/CSS-out; use the cgo path above when you need file-based importers). Or dlopen the dynamic lib at runtime with purego.

Testing & coverage

$ cargo test                                  # unit + integration + doctests (offline)
$ SASSO_PARITY=1 cargo test --test parity # live diff vs dart-sass (needs `npx sass`)
$ cargo llvm-cov --workspace                  # coverage report
$ cargo clippy --all-targets -- -D warnings
$ cargo fmt --check

Changelog

Notable changes are recorded in CHANGELOG.md.

Code of Conduct

As a Sass implementation, sasso adopts the Sass Community Guidelines — see CODE_OF_CONDUCT.md.

License

Licensed under either of Apache-2.0 or MIT at your option.