Sifaka is an open-source framework that adds reflection and reliability to large language model (LLM) applications.

A framework that makes AI research transparent, traceable, and independently verifiable.

BLUX-cA — Clarity Agent core of the BLUX ecosystem. A constitutional, audit-driven AI helm that interprets intent, enforces governance, and routes tasks safely across models, tools, and agents.

A governed system for translating applied AI research into auditable, decision-ready artifacts.

omphalOS turns strategic trade-and-technology analyses into tamper-evident, "run packets" for inspector, counsel, and oversight review.

Self-auditing governance framework that turns contradictions into verifiable, adaptive intelligence.

Deterministic, auditable ethical decision engine implementing the Sovereign Ethics Algebra (SEA).

Cairn is a local-first, cross-platform developer workbench for secure, auditable, AI-assisted software development, helping with coding, debugging, auditing, and collaboration while keeping all projects private and under user control.

🔥 Emergent intelligence in autonomous trading agents through evolutionary algorithms. Testing zero-knowledge learning in cryptocurrency markets. Where intelligence emerges, not designed.

Reference implementation of the Spiral–HDAG–Coupling architecture. It combines a verifiable ledger, a tensor-based Hyperdimensional DAG, and Time Information Crystals to provide a new kind of memory layer for Machine Learning. With integrated Zero-Knowledge ML, the system enables trustworthy, auditable, and privacy-preserving AI pipelines.

Agentic RFP response system orchestrating Sales, Technical, and Pricing agents with human-in-the-loop governance for fast, auditable enterprise responses.

Clinical + Genomic **RAG evaluation (pro)** with hybrid retrieval (BM25 + embeddings), stronger faithfulness, YAML configs, and HTML dashboards. Python **3.10+**.

This repository defines the protocols for **Helix-TTD Identity & Custody**. It enforces a strict "No Orphaned Agents" policy by binding every AI agent to a cryptographic root held by a human custodian.

完全ローカル×思考可視化×三権分立の軽量AIコア。ネット/外部DBなし。state保存/復元、4D意味距離(geo/strict)、署名付き役割カード、𝒢実装。

PULSE: deterministic release gates for AI safety

Ethical AI governance framework for multi-model alignment, integrity, and enterprise oversight.

AILEE Trust Layer is a deterministic trust and safety middleware for AI systems. It mediates uncertainty using confidence thresholds, contextual grace, peer consensus, and fail-safe fallback—transforming raw model outputs into auditable, reliable decisions.

Open, verifiable agent framework — fail-closed orchestration, evidence-based verification, tamper-evident audit chains (SHA-256/HMAC).

This repository provides a compact RAG evaluation harness tailored for clinical + genomic use cases. It operates on de-identified synthetic notes and curated genomic snippets, measures retrieval quality and grounding/faithfulness, and produces JSONL/CSV/HTML reports.