Browse Kotlin Multiplatform libraries

Efficient, generic serialization/deserialization framework with derive-style codegen, pluggable data-format adapters, and faithful line-by-line transliteration including per-file provenance headers, ensuring API parity with upstream implementation.

High-performance lexer generation using compile-time DFA construction, combined token definitions, lookup/jump-table optimizations, regex and literal tokens, loop unrolling, and no-backtracking.

Generate TypeScript declarations from Rust types, including enums, generics and serde-aware mappings; export bindings to files with import resolution and configurable large-integer handling.

Regular-expression engine with deterministic worst-case performance, avoids backreferences/look-around, supports named captures, byte-oriented APIs, RegexSet multi-pattern scanning, iterators, and exposed internals/parser.

Line-by-line transliteration of an upstream identity tool, exposing system identity info (user, host, OS) with a native API, feature-parity, provenance headers and porting discipline.

Loads and parses .env-style key-value files with non-modifying and environment-modifying APIs, configurable loader, reader support, multiline/substitution handling, and compile-time/runtime loading options.

Line-by-line transliteration of a battle-tested base64 implementation offering high-performance encoders/decoders, engine-level APIs, canonical-padding handling, and configurable decoding for non-canonical inputs.

Handles push and local notifications via Firebase Cloud Messaging with simple initialization, configurable foreground display, image support, runtime permission helpers and unified API.

Opinionated error handling & resilience for Kotlin Multiplatform apps using Ktor

HTTP traffic inspector logging requests, responses, headers and bodies; supports mocking/overriding, HAR export, in-app UI, configurable filters, sanitization, retention and notifications.

Runs as an MCP server over stdio, implementing audio processing, subtitle‑to‑LRC and WAV→MP3 conversion, shell command execution, image understanding via Vision API, plus resource/prompt declarations.

Manages runtime permissions, prompts and status via activity-bound controller; coroutine-friendly API, usable isGranted checks outside setup, and lifecycle-aware request handling.

High-performance bindings to libschnorr256k1 offering optimized secp256k1 BIP-340/Nostr APIs: key generation, Schnorr signing/verification (fast and batched), ECDH, SHA-256, automatic native loader.

Build typed property views over raw object maps, decode on-demand, edit known fields, and re-emit preserving unknown fields. Supports JSON/YAML delegates, slices and null-write modes.

Line-by-line transliteration of Starlark syntax: parser, AST and tooling matching upstream Rust, with provenance headers, AST-distance tracking and strict porting discipline.

Run offline LLMs with llama.cpp backend: real‑time token streaming, SHA‑256‑verified resumable downloads, chat templates, KV‑cache reuse across turns, and grammar‑constrained generation.

High-energy UI component library offering 27 neon-styled components, kinetic interactions, bold typography, and a Kinetic Pulse design system with glow accents, squircle shapes, and charts.

Parses .env documents into ordered key/value entries with optional POSIX-style variable expansion, bounded command substitution, preserved quoting/escapes, and precise, line-aware error reporting.

Compact immutable collections, mutable primitive buffers, persistent queues and heaps, plus Option/Either/Validated result types—primitive buffers avoid boxing; includes Arr, ArrMap, BankersQueue, LeftistHeap.

Declarative Gen<T> DSL for property-based data generation, deterministic tape replay, simple shrinking/minimization, generated function values via CoGen, and low-level tape/bit utilities for deterministic testing.

Compact, sorted-range representation of character sets enabling fast membership checks, set algebra (union/intersect/difference/invert), and deterministic partitioning via topology-based refinement for parsers, lexers, automata.

Derives OpenAI function-tool schemas from serialization descriptors, turning sealed request hierarchies into SDK tool declarations and decoding model function calls into typed request objects.

Lightweight toolkit for Ktor HTTP clients: retry schedules with Retry-After parsing, per-request timeouts, etiquette headers, and destructive body-prefix sampling for diagnostics.

Maps structured serialization models to deterministic SCREAMING_SNAKE_CASE environment variables and back into typed configs, supporting nested objects, indexed lists, encode-defaults control, and testable map-based API.

Cursor-based character input layer for parsers, offering lookahead, mark/reset, source positions, span capture, in-memory and streaming inputs, plus small default token parsers (identifiers, quoted strings, whitespace, decimals).

Deterministic pseudo-random generators with mutable Random implementations and serializable immutable-state snapshots for reproducible simulation, testing, procedural generation; includes Philox, Threefry, xoshiro and L64X128 with block APIs.

Lazy, composable, memoized computations that evaluate on demand and cache results; supports map/flatMap/zip, recursive memoization builders, Delay abstraction, and stack-safe composition via internal trampoline.

Identity-based reference wrapper and collection helpers enabling reference-equality keys, identity sets/maps, lazy distinct-by-identity sequences, and diagnostic identity strings for interning, memoization, and cycle detection.

Dependency-light utilities rendering Java-style escaped strings and characters with configurable quote handling, char-class friendly escapes for [] and -, Unicode/control escape forms, and truncation with suffix.