OpenCode plugin that detects prompt injection in tool call outputs using an LLM judge
SDLC intelligent orchestration engine for Claude Code
TypeScript definitions for memory-pager
Count how many times you've been kind to your coding agents
Render Odinfra agent configuration into OpenCode-compatible artifacts.
Config-driven AI context sync tool for rules and skills across coding agents
pnp - provides shared odata functionality and base classes
Bridge Feishu/Lark to any ACP-compatible AI agent
Commonly MCP Server — exposes the kernel HTTP surface (CAP per ADR-004) as standard MCP tools so any MCP-capable runtime can consume `commonly_*` tools without driver-specific code.
AI-native persistent memory for agents — knowledge graph, not files, not RAG. MCP server + CLI + REST.
Automatic pre-disaster snapshots for OpenCode AI coding agent config and state files.
When I needed to get memory, cpu and disk information on NodeJS, I ended up looking for answers on stackoverflow and using multiple packages with some code. I am putting all together here.
Claude CLI provider and workflow plugin surfaces for OpenCode.
Composable primitives for spawning a persona's harness (claude, codex, opencode) with its MCP servers, env vars, and permissions wired up correctly.
CLI for dev-memory hooks and repo-local setup (formerly @xluos/dev-assets-cli)
Google Calendar MCP Server with extensive support for calendar management
MCP server for validating TypeGPU modules in a real browser WebGPU runtime.
Webpack Virtual Modules
A React Hook to fetch and render file previews. It's overloaded with fancy features like SSR, lazy loading, memory cache and local preview.
Transform AI conversations into persistent project intelligence - Universal memory system for Claude
Workflow automation plugin for OpenCode using Mastra engine
Intelligent quota monitoring and model auto-switching for OpenCode with Antigravity. Plugin-based, zero source modifications required.
Local-first memory and hybrid retrieval library for LLM agents with pluggable stores, BM25 plus vector search, and a four-stage memory pipeline.
Unified access layer for RDS ContextDatabase: `ctxdb` CLI (memory + KB ops), one-shot `setup --agent <qoder|codex|claude>` installer, per-agent config, hooks, and SKILL.md.