Hypatia — Security skill for Claude Code
Composable AI agent skills that turn requirements into a traceable, technology-neutral system design — requirements, drivers, architecture style, diagrams, integration contracts, security and reliabil.
How to install Hypatia
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open whitedwarf7/Hypatia and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Hypatia does
Composable AI agent skills that turn requirements into a traceable, technology-neutral system design — requirements, drivers, architecture style, diagrams, integration contracts, security and reliability reviews, and ADRs. Works with GitHub Copilot, Claude Code, and any Agent Skills-compatible agent.
Alternatives in Security
- Pal — Multi-model AI integration — chat, debugging, code review, planning, security audit 11.3k ★
- Query Token Audit — Audit token security to detect scams, honeypots, and malicious contracts across BSC, Base, Solana, and Ethereu 483 ★
- AWS MCP Server — by alexei-led - Features multiple Python environment setup options with detailed code style guidelines, compre 182 ★
README
Hypatia
Agent skills for system design. Give an AI coding agent a product idea or a set of requirements, and get back a traceable, technology-neutral architecture: requirements, drivers, a justified style, a high-level design, diagrams, integration contracts, security and reliability reviews, decision records, and a final review.
Everything lives in [system-design-skills/](system-design-skills/).
Quick start
You do not need to clone this repository. From the root of the project you want the skills in:
npx github:whitedwarf7/Hypatia --target github
That installs all fourteen skills into `.github/skills/` of the current directory, with their shared references bundled so each skill is self-contained. Add `--dry-run` first if you want to see the plan.
If you have cloned this repository, the same installer runs locally:
cd system-design-skills
node scripts/install-skills.mjs --dir /path/to/your/project --target github
| Target | Location | Agents |
|---|---|---|
github |
.github/skills/ |
GitHub Copilot in VS Code, Copilot CLI, Copilot coding agent |
agents |
.agents/skills/ |
Agent Skills standard consumers |
claude |
.claude/skills/ |
Claude Code and compatible agents |
user-copilot, user-agents, user-claude |
~/.copilot, ~/.agents, ~/.claude |
The same agents, for every workspace |
Add `--dry-run` to preview, `--update` to refresh an earlier install, `--skill a,b` for a subset, and `--list-targets` to print the full list. The installer needs Node.js 22.13+ and never overwrites a directory it did not create. Installed skill directories are ignored by git, so install wherever you like without dirtying the repository.
`npx github:` uses git under the hood to fetch into npm's cache, never into your project. Without git, download the repository archive from GitHub and run `bin/install-skills.mjs` from the extracted folder instead.
What you can ask
Start anywhere. Ask for a whole design, or f
Related Skills
Check
by rygwdn - Performs comprehensive code quality and security checks, featuring static analysis integration, se
Eth To Sol Skill
AI skill for converting existing Ethereum contracts into Solana programs, while benefitting from Solana archit
Deep
Execute the current task using DEEP mode for architecture, security, data, infrastructure, public contracts, o
Agenfk Review
Perform a deep code review for security, requirements, and architecture
UI Generate
Full UI pipeline: architecture plan → user approval → implementation → audit. Provide feature requirements as
Arckit.Agent Governance
Design AI agent governance — oversight models, approval workflows, audit requirements, compliance mapping
Related Agents
Timps ADR Writer
Turn a technical decision into a Michael Nygard-style Architecture Decision Record: context, options matrix, d
Integration Explorer
Identifies extension points, existing tests, integration surfaces, and API contracts that a new feature must r
Requirements Architect
Analyze ambiguous feature or bug requests and produce traceable requirements, acceptance criteria, invariants,