Apache 2 Python Packages

predacore

PredaCore — the apex autonomous agent. Hybrid Rust memory kernel, topped LongMemEval R@5 = 0.9574.

signet-sign

Capability-based safety gates for LLM agents. The model proposes; signet authorizes.

alphadynamics

Compact sequence-only neural propagator for protein torsion dynamics. 2.39x lower JSD than Microsoft Timewarp at 3000x fewer parameters. pip install alphadynamics.

sentinel-kernel

Agility infrastructure for regulated AI. Trace, Policy, Evidence under EU jurisdiction. Apache 2.0, on-premise capable.

diarize

Speaker diarization for Python — "who spoke when?" CPU-only, no API keys, Apache 2.0. ~10.8% DER on VoxConverse, 8x faster than real-time.

singleaxis-fabric

Audit-ready OSS substrate for AI agents — inline guardrails, telemetry, LLM-as-judge, escalation. Apache-2.0.

matyan-api-models

Open-source ML experiment tracking system built on FoundationDB — a self-hosted alternative to Aim/AimHub with a Kafka-based ingestion pipeline and cloud artifact storage.

agentboot-ai

AgentBoot — AI agent for bare-metal hardware detection and OS installation. Conversational sysadmin from your phone. Part of P2PCLAW ecosystem.

holographic-reservoir

Holographic Reservoir Computing Framework (Simulation + Optional ASIC Hardware)

autoresearch-nano

nanoGPT-based autonomous ML research loop (research prototype).

quantum-gps-navigator

Quantum-Inspired GPS Navigator: directional-diffusion GPU pathfinder using a 4-channel (N/E/S/W) amplitude metaphor. 100% classical.

arc2-chimera

Research proof-of-concept. ~2-5% accuracy on ARC-AGI. NOT an ARC Prize submission. GPU compute-shader playground.

pinion-mcp

AI-powered characterization test generator for Python — MCP server + CLI. Pin legacy code behavior in pytest before you refactor.

qesn-mabe

Quantum-inspired Echo State Network (QESN) — classical energy-diffusion reservoir with complex amplitudes.

german-ocr

High-performance German document OCR - Local & Cloud with GPU/CPU support

json-correction-loop

Domain-neutral critic→plan→execute loop for large LLM-generated JSON state. Surgical RFC 6902 patching with path_finder + context narrowing sub-agents. At 100 entities full-regen breaks (0% fix); this stack stays at 100% with 5× fewer tokens.

ies-ingest

Orchestration and contract layer for IES-shaped state-demand ingestion. Apache 2.0 donation to the India Energy Stack Accelerator track. Maintained by India Energy Atlas (energymap.in).

geminilens

Local-first observability for Vertex AI Gemini agents: traces, USD cost, drift, tool egress audit. Apache-2.0.

matyan-frontier

Open-source ML experiment tracking system built on FoundationDB — a self-hosted alternative to Aim/AimHub with a Kafka-based ingestion pipeline and cloud artifact storage.

matyan-ui

Open-source ML experiment tracking system built on FoundationDB — a self-hosted alternative to Aim/AimHub with a Kafka-based ingestion pipeline and cloud artifact storage.

maxwells-defense

Adaptive proof-of-work defense for AI agents. Asymmetric thermodynamic cost: attackers dissipate, defenders verify in O(1). Reference implementation of T-IB-09 (Adversarial Dissipation Theorem) — Aristotle-verified. Apache-2.0.

matyan-backend

Open-source ML experiment tracking system built on FoundationDB — a self-hosted alternative to Aim/AimHub with a Kafka-based ingestion pipeline and cloud artifact storage.

matyan-client

Open-source ML experiment tracking system built on FoundationDB — a self-hosted alternative to Aim/AimHub with a Kafka-based ingestion pipeline and cloud artifact storage.