claude-system
Deterministic workflow enforcement, policy gates, and code review loops
All Skills in This Repository (7)
Pure Emerald Level Indicatorsdecide
Generate interactive decision configurators from research, plans, or conversation context.
deep-research
Query OpenAI, Perplexity, and Gemini in parallel and synthesize comparative research reports.
uplevel
Audit repository health across security, documentation, code quality, testing, and standards.
prd
Generate structured PRDs with requirements, metrics, and acceptance criteria.
consume-content
Generate traceable content snapshots with verbatim quotes and structural outlines.
context-preservation
Generate structured context summaries for AI session continuity.
generate-paper-snapshot
Generate PaperSnapshot JSON with verbatim quotes and structural analysis.
Frequently Asked Questions
FAQPage SchemaHow to install claude-system?โผ
Run `npx skills add juanandresgs/claude-system --all -g -y` in your terminal to install all skills globally. Note the upstream repository is marked deprecated, so use it as a reference rather than an active dependency.
What problem does claude-system solve?โผ
It stops coding agents from forgetting rules mid-session by enforcing policies mechanically through hooks and a runtime, instead of relying on prompt instructions alone.
How does the planner-implementer-reviewer loop work?โผ
A planner defines scope, an implementer writes code inside that scope, a read-only critic reviews it to convergence, and a guardian lands the change only after tests and scope checks pass.
Does claude-system work with Claude Code and Codex?โผ
Yes. It is built as a Claude Code configuration with hooks and skills, and it integrates Codex CLI as a read-only review critic inside the workflow.
Is claude-system still maintained?โผ
No. The author has marked the repository deprecated and unmaintained, so it is best used as a historical reference for deterministic agent governance patterns.
Related Repositories in Software Engineering
View All in Software Engineeringโopenclaw
Run a personal AI assistant across your devices and chat apps
superpowers
Gives coding agents a disciplined workflow from idea to merged code
react
AI agent skills for building, testing, and porting React core