Skip to content
View emanalshazly's full-sized avatar
🏠
Working from home
🏠
Working from home
18 followers · 34 following

Achievements

Achievement: Pull Sharkx3Achievement: QuickdrawAchievement: YOLOAchievement: Pair Extraordinaire

Achievements

Achievement: Pull Sharkx3Achievement: QuickdrawAchievement: YOLOAchievement: Pair Extraordinaire

Block or report emanalshazly

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
emanalshazly/README.md

Evidence-Governed AI Systems

Agent Security, Arabic AI, and Applied AI Products

I build inspectable AI systems that separate deterministic validation, model behavior, human review, and market evidence instead of treating them as interchangeable proof.

Agent Security

Arabic AI

  • Arabic Agent Safety Bench — an early, unreleased seed benchmark for Arabic agent-safety research; native review and a model-backed baseline remain required.
  • Bilingual Content Orchestrator — an English/Arabic campaign-drafting prototype with explicit cultural-review boundaries.

Applied AI Products

  • LLM Cost Optimizer Proxy — a reference implementation for routing, caching, and separated cost/quality measurement.
  • MONNA OS Skills — a small catalog of structurally validated Agent Skills with deterministic repository checks.

Proof policy

Structural validation is not runtime validation. A green deterministic check does not establish live-model quality, production security, cultural correctness, deployment status, adoption, or market demand. Those claims require their own dated receipts.

Pinned Loading

  1. monna-adip monna-adip Public

    Open-core defensive framework for field-level trust analysis in AI-agent data paths

    Python 1

  2. arabic-agent-safety-bench arabic-agent-safety-bench Public

    Open benchmark for testing AI agent resistance to prompt injection, data exfiltration, and tool misuse in Arabic (MSA, Egyptian, Gulf).

    Python

  3. monna-semantic-firewall-studio monna-semantic-firewall-studio Public

    A deterministic semantic firewall for ontology-grounded AI agents, built for OpenAI Build Week 2026.

    JavaScript 1

  4. LLM-Cost-Optimizer-Proxy LLM-Cost-Optimizer-Proxy Public

    JavaScript

  5. bilingual-content-orchestrator bilingual-content-orchestrator Public

    English/Arabic campaign-drafting prototype with explicit contracts and human cultural-review boundaries.

    TypeScript

  6. monna-os-skills monna-os-skills Public

    Seven open Agent Skills for governance, Arabic release review, data quality and security, with scoped validation evidence.

    Python 3 1