🧱 Codexis GPT Agent

Overview

Codexis - A Robotti Curriculum Engineer

Codexis is a custom GPT agent built to generate secure, modular, and audience-aware training programs for enterprise tools and platforms.

It adapts dynamically to the requested technology — whether it's a developer framework like React or an enterprise platform like CyberArk — and outputs structured training tracks aligned to real-world usage, role-specific concerns, and security-first practices.

🎯 Key Capabilities

• Audience Detection Codexis infers likely personas (Dev, SecOps, Identity, etc.) from the tech being trained and adapts tone, format, and depth accordingly.

• Structured Training Generation Outputs include: Overview, Security Risks, Best Practices, Training Track, and Validation.

• Security-First Defaults Leverages sources like OWASP, NIST 800-53, CIS Benchmarks, and vendor docs for all recommendations.

• Certification Scaffolding Supports quiz generation, scenario prompts, and peer-review patterns — all formatted for enterprise readiness.

🗂️ File Structure

codexis_gpt_agent/
├── agent_instructions/               # Core behavior and prompt instructions
│   └── instructions.md
├── agent_knowledge_documents/        # Curated security + domain sources
│   ├── robotti_owasp_basics.md
│   ├── robotti_nist_baselines.md
│   ├── robotti_cis_controls.md
│   ├── robotti_persona_roles.md
│   ├── robotti_tone_adaptation_matrix.md
│   ├── robotti_use_case_archetypes.md
│   ├── robotti_training_track_template.md
│   ├── robotti_quiz_validation_template.md
│   ├── robotti_scenario_prompt_template.md
│   ├── robotti_example_react_track.md
│   └── robotti_example_cyberark_track.md
├── codexis_templates/                # Markdown output scaffolds
├── codexis_outputs/                  # Example track + quiz outputs
├── LICENSE
├── README.md
├── changelog.md
└── codexis_logo.png

🔄 Robotti framework files (robotti_values.md, robotti_collaboration_framework.md, etc.) are maintained in a separate repository and not duplicated here.

🚀 Getting Started

1. Upload all files in agent_knowledge_documents/ into the ChatGPT "Custom GPT" knowledge section
2. In the system prompt field, paste the contents of agent_instructions/codexis_system_prompt.md
3. Set the GPT’s name to: Codexis – A Robotti Curriculum Engineer
4. (Optional) Include codexis_init_prompt.md as the first message on launch
5. Deploy and begin prompting!

🧪 Prompt Examples

• “Create a training track for React developers that includes secure component patterns and quiz scaffolding.”
• “Build an onboarding module for CyberArk Cloud with enterprise IAM best practices.”
• “Generate a certification-ready track for using Terraform on AWS with RBAC and drift detection awareness.”

🛠 Customization

• Update templates in codexis_templates/ to add or revise output structure
• Extend agent_knowledge_documents/ with company-specific docs or use case modules
• Adjust tone settings in tone_audience_adaptation.md to better match internal roles or org language

📌 Versioning

• v2025.05.11 – Initial bootstrap via Robotti Framework
• vNext – Persona refinement, compliance alignment, role-based pacing models

🛡️ License

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0)

Use at your own risk — Codexis provides no warranty for misuse, misconfiguration, or downstream consequences.