verifiable-ai

OML 1.0 via Fingerprinting: Open, Monetizable, and Loyal AI

~97% token reduction for AI coding sessions — zero deps, 33 languages, MCP server

The agent harness that makes small local models finish hard tasks — and prove it. Done is enforced by the engine, not the prompt: gates, rewinds, a judge, and a replayable Proof-of-Done receipt on every finished run. Any model in the socket. Local, sovereign, air-gapped-capable. macOS · Linux · Windows.

AI-powered crypto trading signals verified in OpenGradient TEE. 6 technical indicators + LLM macro-risk verdict + ONNX volatility forecast. Next.js 16 · React 19 · TypeScript · Tailwind CSS 4 · Binance API · onnxruntime

Offline, on-device AI desktop agent with a signed, independently-verifiable audit trail — runs sealed with zero verified network egress; every action is hash-chained and checkable offline. MIT · pre-launch.

A commit-and-audit proof system for deterministic, quantized inference of a JEPA-style world model (LeWorldModel)

The verifiable memory cortex for AI agents: a semantic graph and vector search with zero-knowledge provenance.

The frozen cryptographic kernel of the AXM protocol. Signed, content-addressed knowledge shards with post-quantum signatures (ML-DSA-44) and mechanically enforced non-selective recording. The gold shard is the definition of correctness.

FDE Consultants Protocoles — open-source Forward Deployed Engineer + DeepSCR protocol: turn any coding agent (Claude Code, Codex, Cursor) into a certifying engineer with verifiable AI Assurance Scores, MCP tools, and a public trust registry. Apache-2.0.

AI-powered GitHub repo security analysis, verified on-chain via OpenGradient TEE

Run AI agents locally with built-in governance and verifiable proof. Loom + Kernel + Intelligence.

This unified document integrates glyph specification, dictionary, error taxonomy, canonical taxonomy, grammar, resonance matrix, security/custody, and operational runbooks into one continuous, human-centric format.

The open standard for AI agent integrity. Evaluate, enforce, and prove that autonomous agents are adversarially coherent, environmentally portable, and verifiably assured.

VELLA = Verifiable Evidence Ledger & Logic Authority; deterministic pre-execution authorization and signed proof for AI agents and automated workflows.

Canonical monorepo for the Aethelred protocol, security baselines, and public release provenance.

Public hub for the Architected Instruction & Evidence Protocol — specifications, schemas, governance, and machine-readable discovery endpoints. Apache-2.0.

TriAgentFlow (TAF) — a three-role (Leader / Worker / Verifier) AI development workflow with plan-grounded audit governance. Formerly HTE.

Thirteen autonomous AI agent demos on Base Sepolia. Each reproduces a real-world failure (Mango oracle, Terra peg, Ronin bridge, MCAS cert) and lets the agent refuse — every decision signed on-chain with the dossier published. Eight adjudication agents + five non-adjudication (NFT authors, critic, legal co-author, in-game NPC).

Pi extension that records every agent run as a verifiable proofline bundle

Sovereign Enterprise Agent Fleet — Fortified Enterprise Fleet track (#AllThingsAgenticHackathon). Local-first authority, cryptographically verifiable on GCP.