Generate STRIDE threat models for system architectures. Identifies Spoofing, Tampering, Repudiation, Information Disclosure, Denial of Service, and Elevation of Privilege threats with risk ratings, CWE references, and actionable mitigations.

Commit and push only files modified by the AI agent across the current session. Scans the conversation transcript to identify edited files and cross-references with git status to detect prior session changes.

HTML-native design skill with fact-check protocol, brand-asset workflow, junior-designer iteration, philosophy advisor, anti-slop checklist, and 5-dimension expert review. Adds ~6KB to the agent system prompt via the inline_prompt mechanism; references load on-demand via read_file.