-
GNAT Limited
- New Zealand
- http://nathan.torkington.com
Starred repositories
A local-first personal CFO for a variable-income founder. Pure decision layer, Postgres, Streamlit, Tiller ingest.
TencentDB Agent Memory is a team-level memory hub for AI Agents — turning conversations, docs, and code into four reusable memory assets (Chat Memory, Skill, LLM-Wiki, Code-Graph) that are governed…
The CLI for AI agents to control Chrome. Zero config, agent-agnostic, battle-tested.
Easy isolated self-hosted GitHub Actions runners (VM-preferred: Tart/libvirt, Docker fallback with hard limits)
Turn a codebase, workflow, or prompt pipeline into a polished, interactive HTML flowchart — a skill for AI coding agents.
Dart framework for stateful AI agents: tool use, skills, sub-agent delegation, planning, streaming, evals, and multi-provider LLM support.
🪄 Flint is a visualization language that lets AI agents reliably create expressive, good-looking charts from simple, human-editable chart specs.
A maintainability ratchet for AI-assisted Python.
Confucius4-TTS: a Multilingual and Cross-Lingual Zero-Shot TTS Engine
Unlimited OCR Works: Welcome the Era of One-shot Long-horizon Parsing.
A real-time order analytics pipeline built on CDC from Postgres to ClickHouse on the Brazilian Ecommerce Dataset
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
HEPA: Self-supervised horizon-conditioned event predictive architecture for time series. Spotlight at FMSD @ ICML 2026.
Transparent proxy server that works as a poor man's VPN. Forwards over ssh. Doesn't require admin. Works with Linux and MacOS. Supports DNS tunneling.
Agent skill for overcommenting audit copies with dependency traces and PASS scoring
Audio Plugin for Audio to MIDI transcription using deep learning.
Execution-Layer Security (ELS) for AI agents — policy-enforced shell with audit.
Zero-dep agent workflows with structured approval gates and adversarial review.
Multi-lens code review pipeline for Claude Code: deep review (Claude or Codex), auto-fix loop, interactive walkthrough, external-finding injection.
WorksGood: work coordination graph for humans and AI agents
A spatial design tool where everything renders as Unicode box-drawing characters. Create wireframes, diagrams, and mockups. Share as text
Local-first session search, analytics, insights, and token use statistics for coding agents, supporting Claude Code, Codex, and more than 20 other agents.
A non-interactive coding agent: give it a prompt and it reads, writes, runs commands, and searches code in a loop until the work is done, using native tool-calling across OpenAI, Anthropic, and Goo…
Agency: a primordial soup of primitives for making machine organisations
karpathy auto research ported to mlx so you can use your Mac