Search

The AI assistant for your product: drop it into your React app or any page with a script tag.

Sudarshana — Premium Futuristic Spiritual AI Assistant inspired by a cosmic universe

v2-native secret detector/redactor: scrubs the outbound HTTP body (title/compaction/generate), prompt, tool args/results and child-process env; observe/redact/block modes with entropy detection, allowlist precedence and a hashed JSONL audit.

A tiny Pi extension that shows how much time and model usage each prompt consumes.

Route OpenCode tool selection through TypeSafe System-1 (Jev): a cheap fast model picks the next tool so the reasoning model executes instead of deliberating.

Toolkit for agent-assisted NEXT campaign builds.

Listens to an audio file and answers with a voice, through a speech account the owner chose, with daily caps.

On-device (and optionally cloud) ML inference for React Native: LLM, ASR, TTS, embeddings and vision, backed by the Xybrid Rust SDK

Sebas, the electronic butler for LLM-assisted work: speaks your agents' messages through a local text-to-speech daemon, parks the messages that collide onto a system notification card with a single Listen button, and keeps one voice at a time. Ships the v

Echo 窗口编排插件：作为本机的窗口执行器运行 Echo 窗口工位（切换会话、粘贴发送、停止），并支持 OCR 触发词自动点击、动作编排、执行记录与日志。

Scaffolder for AI Launch Kit — `pnpm create ailk@latest <dir>` produces a starter repo consuming @working-theory/* from npm (per ADR 0007).

Shared contracts (types + zod schemas) for DevEye apps

MetriBeacon Agent and CLI for Linux, macOS and Windows

MetriBeacon Agent and CLI for Linux, macOS and Windows

MetriBeacon Agent and CLI for Linux, macOS and Windows

MetriBeacon Agent and CLI for Linux, macOS and Windows

What every lakatos-ts tool shares at runtime: the result envelope and its schema, the SZS status vocabulary, interrupt handling, and run-directory claiming.

全景图标签和标注的共享视觉组件。宿主负责坐标投影、数据读取，以及通过 `onClick` 绑定业务行为。

CLI for managing markdown documents with YAML frontmatter — index, query, validate, graph, export, lifecycle, and AI summaries.

Guardrails for the agents you build: permissions, verification, approvals, grounding and budgets from Scute, on the agent framework you already use.