Search

Cron, interval and timeout jobs with distributed locking for Knead, the NestJS-style web framework for Bun

An async dataflow scheduler: nodes that depend on other nodes, run by your async function, with glitch-free propagation, stale-run cancellation, skip values, live nodes, sampled inputs, queued event nodes and per-node run states. Values live wherever you

Pi extension: live log tail inside running shell tool rows, plus a stall/max-runtime watchdog that stops hung builds and summarizes why.

Official headless CMS adapters (Strapi, Contentful, Sanity, Shopify) for next-advanced-sitemap — map CMS records to SitemapEntry[] without writing a mapping loop by hand.

One command DevOps: npx wizard that installs GitHub-native Release Automation (optional GitHub Copilot AI summaries) into any project

Timed, hands-on mock exams for Kubernetes administrator certification prep, on real nodes. Coming soon.

Model 3D game assets as code with three.js and an AI coding agent: one TypeScript recipe per model, built in node into a GLB, checked against your project's budgets and palette, rendered in a headless browser and reviewed with pins in yours. A Claude Code

OpenAI embedding API wrapper + VantaDB storage — generate embeddings via OpenAI and store/search them in VantaDB

Ollama local embedding API wrapper + VantaDB storage — generate embeddings via Ollama and store/search them in VantaDB

LiteLLM embedding gateway + VantaDB storage — unified embedding provider with local vector search

Temporary package placeholder for staged publishing

轻量浏览器 localStorage / sessionStorage 工具库，支持过期时间、命名空间、自动序列化，零依赖。

n8n nodes for sending and receiving Matrix E2EE gateway messages

n8n community node for the Threads API.

DSH 桌面陪伴挂件：Codex 订阅额度监控、手动程序路径与自定义回复，纯 DOM 实现。

Codon model averaging with CPU/GPU execution and numerical diagnostics

TriTopic: graph-based topic modeling that fuses semantic, lexical and metadata views with consensus Leiden clustering, plus LLM topic interpretation and a research toolkit for reliability, saturation, group comparisons and topic evolution.

LLM provider layer of craftery: AI SDK adapter, model router, fallback, Codex provider, context compaction

Entity storage backed connector for persisted vault key and secret workflows

CollieAi guardrail for the LiteLLM proxy: checks every chat completion against your CollieAi policy, in Monitor mode.