In this article
Security Documentation
Overview
This directory contains security documentation for HVE Core, demonstrating defense-in-depth security practices.
Documents
| Document | Description |
|---|---|
| Branch Protection | Main branch protection requirements and repository controls |
| Security Model | Comprehensive security model and security assurance case |
| Dependency Pinning | Pinning strategies and CI enforcement for all dependency types |
| SBOM Verification | SBOM attestation verification and consumption guide |
| VEX Verification | Download, verify, and interpret the published OpenVEX document |
| Fuzzing | OSSF Scorecard fuzz harness convention and compliance |
| Dangerous Workflow Detection | Hybrid CI control: a homegrown template-injection gate plus the Poutine supply-chain scanner for GitHub Actions workflows |
| Code-Scanning Alert Lifecycle | How code-scanning alerts are detected, blocked, tracked, and resolved without dismissal |
| SECURITY.md | Vulnerability disclosure and reporting process |
Skill Security Models
Skills whose shipped scripts make network requests, handle credentials, parse content from outside the operator's control, write outside .copilot-tracking/ or their declared output location, or run an external program with input derived from those surfaces carry a per-skill STRIDE threat model in a SECURITY.md alongside their SKILL.md. Skills that meet none of those triggers are declared exempt, with a reason, in the skill security classification file described in the security model.
| Skill | Runtime surface | Security model |
|---|---|---|
| jira | REST CLI; environment credentials | SECURITY.md |
| gitlab | REST CLI; environment credentials; git-remote subprocess | SECURITY.md |
| mural (experimental) | REST CLI; embedded MCP server; OAuth token store | SECURITY.md |
| tts-voiceover (experimental) | Azure Speech egress; key/Entra credentials; SSML + PPTX parsing | SECURITY.md |
| accessibility | Arbitrary-URL scan egress; npx @axe-core/cli subprocess | SECURITY.md |
| powerpoint (experimental) | Opt-in author-Python content-extra.py execution; LibreOffice/MuPDF parsing | SECURITY.md |
| video-to-gif (experimental) | Local CLI (bash + PowerShell); FFmpeg/ffprobe subprocess | SECURITY.md |
| copilot-otel-metrics (experimental) | Diff-approved global settings write; loopback OTLP ingest into a container stack; local-API query helpers; generated Azure collector and infrastructure templates | SECURITY.md |
| gh-code-scanning | GitHub code-scanning read via gh CLI subprocess | SECURITY.md |
| customer-card-render (experimental) | Local Python CLI; DT markdown to content.yaml emission | SECURITY.md |
| security-planning | Local Python generator; native TMT validation harness; Windows UI Automation; screenshot capture; overlay and evidence handling | SECURITY.md |
| vex | Local Python gate; untrusted issue-body + OpenVEX doc parsing | SECURITY.md |
| hve-core-installer | Local PowerShell + Bash file copy into a target repository; jq, git, VS Code CLI subprocesses | SECURITY.md |
| hve-demo-material (experimental) | GitHub REST and Pages client with GITHUB_TOKEN; deck and video render writes; node, npm, ffmpeg, git subprocesses | SECURITY.md |
Security Posture
HVE Core is an enterprise prompt engineering framework that:
- Contains no runtime services or user data storage
- Operates as development-time tooling consumed by GitHub Copilot
- Relies on defense-in-depth with 20+ automated security controls
The security model documents:
- 36 threats across STRIDE, AI-specific, and Responsible AI categories
- Security controls mapped to each threat
- MCP server trust analysis
- Quantitative security metrics
- GSN-style assurance argument
Related Resources
- Branch Protection: Repository protection configuration
- MCP Configuration: MCP server setup and trust guidance
- GOVERNANCE.md: Project governance and maintainer roles
🤖 Crafted with precision by ✨Copilot following brilliant human instruction, then carefully refined by our team of discerning human reviewers.