Search

Official Node.js/TypeScript SDK for the FalaAI API - AI-powered call transcription, diagnosis and compliance auditing.

Intel is a customer-deployed, model-agnostic operating layer for company intelligence and processes. It gives people and MCP-capable AI clients the same governed access to two surfaces:

OCR capability seam for DeepSeek Harness: ctx.ocr, the engine contract, the result schema, and the fallback pipeline

BornEngine: native TypeScript game engine compiled by Perry

Deterministic fake provider for ForgeaX intelligence consumers.

Runtime-safe code-first GPU VFX source, asset loading, player intent, and fixed-tick GPU work contracts.

osxsystem-context-engine binary for Windows x64

osxsystem-context-engine binary for Linux x64

osxsystem-context-engine binary for Linux arm64

Vite build integration for PWA Platform.

Drop-in TypeSafeClient replacement backed by LLM APIs

Adaptive reasoning effort for OpenCode V2 with request-local GPT-6 model selection

Bulk AI translation for Strapi v5, with admin-managed providers and models

Roll back, branch, and export any DSH conversation — from the UI or straight from the agent via a message_ops tool. Non-destructive: logs stay append-only and recoverable. 消息回滚/删除/分支/导出四合一，内置 message_ops agent 工具；非破坏语义，日志只追加可恢复。

empty

SRL (SPARQL Rule Language) support for CodeMirror 6, on the SPARQL 1.2 Lezer grammar

Spec-Driven Development (SDD) template. Docs are the single source of truth; code implements docs.

osxsystem-context-engine binary for macOS arm64

Richer model picker for the DeepSeek Harness Web composer: a searchable provider select (with logos, model counts, and key status) above a keyboard-first, fuzzy-searchable model list grouped by provider, with favorites, recents, context-window/vision/reas

Ollama-backed search & fetch provider for the DeepSeek Harness `ctx.web` seam (host half of the web-search-ollama plugin)