Search

Private On-Device Inference Engine — run LLMs locally with zero configuration

OpenAI-compatible HTTP server for fusionkit.

Goose — a columnar storage engine with 10 specialized encodings, roaring-bitmap predicate pushdown, bloom-filter partition skipping, zone-map pruning, probabilistic sketches, and an SQL query advisor. Ships with a PostgreSQL profiler/exporter CLI.

Install the Truora design skill (atelier) into any AI coding harness — Claude Code, Cursor, Codex.

A local OpenTelemetry ingest + TUI viewer for development, backed by SQLite. (Executor fork of @kitlangton/motel)

Modular skill tree compiler and AI rules manager for developer workspaces. Integrates dynamically with Cursor, VS Code, Windsurf, Claude Code, and Antigravity.

TypeScript SDK for Zero — search, fetch, and pay for AI capabilities (x402 + MPP).

Multi-model LLM Council + parallel AI window for Claude Code. Run structured council reviews across Gemini, GPT, DeepSeek and more — or fork a conversation to any model and fold the results back.

Unified engineering-partner workflow for Claude Code — debug, build, refactor, review, post-mortem — with hard gates and mechanical (hook-enforced) verification discipline.

Installer for the orchestrator-harness-skills Claude skill — makes an agent operate as a strict orchestrator built on the agent-harness model (LLM + HARNESS).

Manage Agent Skills for Xcode's Codex and Claude Coding Assistant integrations.

Capture the full trajectory of a coding-agent harness (Claude Code, Codex, Gemini CLI) — request bodies, system prompts, tools, streamed responses, token usage — into a JSONL log and a self-contained HTML viewer, by proxying the agent's API traffic throug

Automyx - Professional AI automation suite with full system control, 2D/3D editing, and real-time voice capabilities

Git AI Context Manager - auto-sync project context to AI coding tools after every commit. No API key. Works offline.

The git-native medium where many minds converge on one living artifact — each refines it through their own AI, toward an owned vision. CLI + agent shim; the HTML is just a view.

Delay Codex prompts with /wait and /steer using a UserPromptSubmit hook.

A CLI and Node.js module for tcsas-miniprogram continuous integration.

Vercel AI SDK tools for the Pelican panel Client API

Polls GoNext cloud API for async local LLM jobs and runs them against Ollama/OpenAI-compatible servers on this Mac

ARHA pure emotion-cognition engine — Phase 1-7 (value-driven emotion, authenticity, continuity, lifetime resonance, cross-persona distinction, autonomous self-state, dialogue climate). Framework-agnostic, zero runtime deps.