#llm-guardrails

AX-first computer use for AI agents on macOS with optional Jev (TypeSafe System One) semantic guards: calibrated target/input judgments before an irreversible action, decisions kept in code. Accessibility-tree targeting, window-scoped input, clipboard-safe paste, read-back verification. Ships a pip CLI, a pi package and a DeepSeek Harness plugin.

Unified open-source security shield for agentic AI systems — defense-in-depth prompt-injection protection (canary tokens, agent-trace alignment audit, tool-call guarding, PII/secret scanning).

Retail exception detection and task reliability engine connecting ERP, WMS, and HRMS with deterministic reconciliation, idempotent task workflows, and guardrailed AI advisory.

Turn vague automation requests into build-ready specs with guardrails. A Claude Code plugin (and Codex kit) that runs a guided intake, gates each item Go or Don't Build, and writes a traceable spec pack.

ISO 9001:2015 / Agile V / W3C PROV-DM Quality Pipeline, Anti-Tamper HMAC Guardrails, MCP Server & Benchmark Suite for Multi-Agent Systems