solatisCommunity·12 Agent Skills Included

claude-config

Structured AI coding workflow with planning, review gates, and quality checks

Enforces plan-before-code workflows for AI-assisted development with quality review gates at every milestone. Stops LLM-generated technical debt by catching duplication, ambiguity, and documentation drift before it compounds. Includes skills for root cause analysis, decision stress-testing, refactoring, prompt optimization, and codebase exploration. Keeps agent context lean with indexed documentation so every task gets exactly the information it needs.

901|134

↗

npx skills add solatis/claude-config --all -g -y

Available:

Tells the AI agent how to invoke each skill's Python script immediately, follow the shared prompt organization rules, and route tasks to the right sub-skill without exploring first.

All Skills in This Repository (12)

Pure Emerald Level Indicators

📦 In Repo

solatis

problem-analysis

Identifies root causes of software problems through a scripted five-phase investigation workflow.

deepthink

Orchestrates multi-step structured reasoning for open-ended analytical questions via a Python workflow script.

codebase-analysis

Orchestrates phased codebase exploration producing evidence-backed findings with file and line references.

arxiv-to-md

Convert arXiv papers from TeX source into clean markdown for LLM consumption.

cc-history

Query and analyze Claude Code conversation history JSONL files using shell commands and jq.

refactor

Analyzes codebases for code smells and technical debt using parallel exploration agents.

planner

Plan multi-step projects into milestones with dependency graphs and review gates.

analyze

Run a six-phase codebase analysis workflow for architecture, security, and quality issues.

doc-sync

Synchronize CLAUDE.md navigation indexes and README.md architecture docs across a repository.

incoherence

Detect contradictions between documentation, specifications, and code.

decision-critic

Stress-test decision reasoning by extracting claims, assumptions, constraints, and judgments.

prompt-optimizer

Optimize Claude Code agent system prompts through phased analysis and human approval.

Community

Basic

Frequently Asked Questions

FAQPage Schema

How to install claude-config?▼

Run `npx skills add solatis/claude-config --all -g -y` in your terminal to install all skills in this suite globally.

How to stop AI-generated code from accumulating technical debt?▼

This suite separates planning from execution and runs quality review gates on every milestone, catching duplication, ambiguity, and drift before mistakes compound.

What skills are included in claude-config?▼

It includes a planner with review gates, deepthink for open-ended reasoning, problem-analysis for root causes, decision-critic for stress-testing choices, refactor for code smells, prompt-engineer, doc-sync, codebase-analysis, arxiv-to-md, and cc-history.

Does claude-config work with Claude Code?▼

Yes. It is built specifically for Claude Code and installs into your .claude directory, where agents load the indexed CLAUDE.md files and skills automatically.

Do I need to change how I write prompts to use it?▼

No. You make plain-English requests like 'use your planner skill to write a plan', and the scripts orchestrate the multi-step workflow, sub-agents, and review loops for you.