Browse agent skills

⚡ B-FAST Context Compression Layer for Large Language Models (LLMs)

Durable Vercel AI SDK adapter: journal every model call and tool execution as a Throughline step.

Reusable app-building UI and helpers for Agent-Native apps.

Project-family orchestration workflow: same migration, ralph single-repo loop, multi-project dispatch.

OpenCode plugin and OpenTUI review CLI for collecting fine-tuning datasets.

OpenCode plugin and OpenTUI review CLI for collecting fine-tuning datasets.

htop for Ollama — curses-based TUI monitor for models, GPU, and Docker containers

The ultimate social media scheduling tool

AI-powered SEO pipeline — content audit, keyword research, internal linking, image injection, content generation, publishing

A personal AI assistant daemon — chat, scheduled jobs, persona system, extensible via skills.

Off-DOM CSS layout engine — Flexbox and CSS Grid box positions in pure synchronous JavaScript, no DOM, WASM, or native binaries

Deterministic commenter for agentic PR review — render, inline, post, adapt, cost, validate, print-schema

Audit whether an AI answer engine (ChatGPT, Perplexity, Google AI) would cite a web page: a two-stage retrieval-vs-quotability model plus an 'is this query even winnable for an independent site?' verdict.

A Claude Code skill: indie-maker checklists inspired by Pieter Levels' MAKE, organized by Idea/Build/Launch/Grow/Monetize/Automate/Exit.

Synod feedback extension core SDK — HTTP client, lifecycle management, and React wrapper for submitting external findings.

Pi package that watches pull requests with PR-aware /vette, /pr, /watch, and /peek commands plus GitHub status diagnostics.

jgengine core — pure-TypeScript game framework SDK (npm @jgengine/core). Entity stores, commands, combat, inventory, multiplayer seams. Zero dependencies. Not an ECS. Agent skills ship in the tarball. https://jgengine.com

Library for using Modern VK Audio API

Open-source model portability and replacement benchmark: measure which model delivers an acceptable outcome at the lowest cost, latency, and operational risk.

On-device semantic embeddings for JavaScript — the small, fast tier. 1.58-bit BitNet-style sentence embedder via WASM. ~5 MB on the wire, ~2.5 ms per embedding, no API calls.

Receipt every tool call in a LangChain or LangGraph run with one callback handler - no changes to your tools. Signed, verifiable Zanii proofs across the whole chain.

AI integration helpers for the jth runtime via Ollama — a **JavaScript-only helper library**, not a jth op package.

Compiler pipeline for the jth language. Transforms jth source code into executable JavaScript through three stages: lexing, parsing, and code generation.

Compose, reuse, and run your own agent workflows (mini teams). Name them, run them on any task, watch each step.

QraftAI local device runner — bridge your own Android emulator/phone to a QraftAI workspace.

Composable, plugin-first library for building LLM agent harnesses. Provides the agent loop, tool registry, session/context management, policy engine with command-safety rules, provider/credentials registries, and an in-process subagent primitive.

Production-grade prompt engineering toolkit. Type-safe, composable, and extensible.

LangChain Tool integration for The Factory — x402 micropayment data API

Circuit SDK core — http client, injectable config, ed25519 identity, shared types.

Defense against LLM Rust failure modes for Claude Code and Codex: 58 categories of bugs that survive rustc, clippy, and cargo test.