~/arthur.pan Projects Writing GitHub

~/writing

Writing

Project write-ups & essays on agentic AI and security.

  • Jul 4, 2026

    DvalinCode: a coding agent you can actually let into a regulated codebase

    An AI coding agent built for regulated environments — org-policy enforcement over tools, models, shell and file paths, sandboxed execution, and a hash-chaine...

    agentic-ai · security · developer-tools
  • Jul 4, 2026

    Building DocSentinel: six agents instead of one big prompt

    Why I built DocSentinel as six specialised agents coordinated by a graph, and what hybrid vector + knowledge-graph retrieval actually buys a security review.

    agentic-ai · security · rag
© 2026 Arthur Pan · built with Jekyll GitHub · LinkedIn · Email