A customized Claude Code fork for routed models, multi-agent orchestration, and dense terminal control.

Python SDK for dataset generation on LightningRod platform ⚡

Self-hosted control plane for AI coding agents (Claude Code, Codex, Gemini, Ollama). Shared MCPs, reusable workflows, AI audit that makes any repo AI-friendly. Engineering, not prompting : smaller prompts, fewer hallucinations, lower token bill.

Template-aware development workflow plugin for Claude Code.

A powerful Model Context Protocol (MCP) server and Omni-AST engine. It empowers AI agents to seamlessly parse complex codebases, execute secure cross-project operations, and dynamically fetch token-optimized rules.

AI agent skills library built by OpenSite AI. A single git repo that keeps Claude, Codex, Cursor, Copilot, and Perplexity Computer in perfect skill sync, with persistent cross-session memory and automated cloud upload via Playwright.

AI-first frontend design toolkit for Claude Code and Codex: 7-stage flow with six-dimension research, dualround interview, visual anchors, motion spec, anti-slop gate, and kun 15-segment DESIGN.md contract. Constraints before generation.

A Claude Code skill that turns every AI request into a clear, constrained and verifiable prompt before execution.

Adaptive wizard that generates production-ready configs for Claude Code, Cursor, GitHub Copilot, Gemini CLI, Windsurf, and AGENTS.md from a single Q&A. Deterministic, offline, audit-ready. compliance-aware (HIPAA / PCI-DSS / SOC2 / GDPR / FERPA / COPPA).

Daily self-improvement loop — scans configured sources for new patterns, diffs against your current setup, implements quick wins, writes a dated report. Originally a Claude Code skill, now config-driven for any ecosystem.

finasys: Fast financial data processing, feature engineering & AI agent toolkit

A cognitive quality harness for Claude Code — hard gates, deep thinking, every domain. No npm. No runtime. Just quality.

A hook-based router for Claude Code: classifies every prompt, dispatches to the right framework, and enforces runtime guardrails.

Single-binary operational CLI for Qdrant with .qql runbooks, CI automation, retrieval diagnostics, and stable JSON output

Fast BPE tokenizer with guaranteed perfect text reconstruction. 2x faster than tiktoken at encoding/decoding while preserving all whitespace and formatting.

HTML as canonical LLM memory and inter-agent comms. Claude Code plugin + MCP server. Anchor-signed, schema-validated, sanitized. Apache-2.0.

Context composition engine for AI systems. Builds structured prompt payloads from Markdown artifacts and renders them for OpenAI, Anthropic, and agent frameworks.

System-level lint for multi-agent harnesses. Catches the 21 structural traps single-file linters miss — including the LLM-when-you-should-use-code patterns that burn tokens.

Lint, test, and score agent instruction files (CLAUDE.md, AGENTS.md, MCP tool specs).

Harmonic-substrate programming language: first-class φ, dual-band LLVM JIT, self-healing compiler, O(log_φπfib N) algorithms, and a substrate-native ML framework whose substrate-aware transformer attention wins -8.94% val on TinyShakespeare.