Skip to content

github_launch_plan.md

Latest commit

 

History

History
67 lines (50 loc) · 2.54 KB

github_launch_plan.md

File metadata and controls

67 lines (50 loc) · 2.54 KB
Raw

GitHub Launch Plan

Repository Positioning

Suggested repo name:

fde-ai-systems-portfolio

Suggested tagline:

Three runnable enterprise AI systems showing secure RAG, governed agents, AI release reliability, evals, traces, audit logs, and approval gates.

Why People Might Star It

  • It is not another chatbot template.
  • It shows three complete enterprise AI systems.
  • It runs locally without paid APIs.
  • It includes evals and smoke tests.
  • It demonstrates security, governance, and reliability patterns reviewers can inspect directly.
  • It includes production upgrade notes.

Launch Checklist

  • Confirm README renders well on GitHub.
  • Add visual README assets.
  • Replace static quality badge with GitHub Actions badge after publishing.
  • Add screenshots.
  • Add short README GIF.
  • Run all evals and smoke tests.
  • Add repo topics:
    • ai-agents
    • rag
    • llm-evals
    • forward-deployed-engineering
    • enterprise-ai
    • openai
    • agentic-workflows
  • Apply repository settings from docs/github_repository_settings.md.
  • Run python -B scripts/post_publish_check.py.
  • Convert docs/community_backlog.md into the first public issue wave.
  • Create and complete the first public issues, then refresh docs/github_initial_issues.md into the current community issue pack.
  • Record the first lightweight demo GIF using docs/demo_recording_checklist.md.
  • Add a deterministic launch asset hygiene gate through python -B scripts/dev.py launch-assets.
  • Publish launch copy from docs/launch_copy_pack.md.
  • Publish a short technical post.
  • Pin the repo on GitHub profile.

README Hook

Lead with the problem:

Most AI app demos stop at chat. Real enterprise deployments need permissions, citations, approval gates, release gates, audit logs, traces, and evals. This repo implements those patterns in three runnable systems.

Launch Post Draft

Open-source reference systems for enterprise AI controls:

  • Secure Enterprise Knowledge Copilot: permission-aware RAG with citations, abstention, prompt-injection handling, traces, audit logs, and evals.
  • Regulated Customer Operations Agent: governed tool-calling workflow with approval queues, side-effect blocking, supervisor approval, audit logs, traces, and unsafe-action evals.

It runs locally without paid APIs, but includes optional OpenAI Responses API integration points and a production upgrade path.

The goal is to show what AI apps need after the chatbot stage: governance, observability, and quality gates.