ai-agent-stack skill

A Claude Code skill that adds production safety to any AI agent codebase in minutes.

What it does

Installs and integrates three zero-dependency Python libraries:

• ai-cost-guard — hard budget cap before LLM calls (pip install ai-cost-guard)
• ai-injection-guard — prompt injection scanner (pip install ai-injection-guard)
• ai-decision-tracer — local agent decision tracer (pip install ai-decision-tracer)

Install in Claude Code

/plugin marketplace add LuciferForge/lucifer-skills

Or install directly:

/plugin install ai-agent-stack@lucifer-skills

Usage

Just describe what you want:

• "Add cost protection to my agent"
• "Instrument my LLM calls with tracing"
• "Protect my bot from prompt injection"
• "Make my agent production-ready"
• "Add all three safety layers"

Claude will analyze your codebase, find the LLM call points, and wrap them with the right guards.

The libraries

Library	GitHub	PyPI
ai-cost-guard	github.com/LuciferForge/ai-cost-guard	pip install ai-cost-guard
ai-injection-guard	github.com/LuciferForge/prompt-shield	pip install ai-injection-guard
ai-decision-tracer	github.com/LuciferForge/ai-trace	pip install ai-decision-tracer

All three: MIT licensed, zero dependencies, pure Python stdlib, no network calls.

License

Apache-2.0