@ai2070/l0
L0: The Missing Reliability Substrate for AI
llmverify
AI Output Verification Toolkit — Local-first LLM safety, hallucination detection, PII redaction, prompt injection defense, and runtime monitoring. Zero telemetry. OWASP LLM Top 10 aligned.
llm-inject-scan
A tiny, fast library that scans user prompts for risky patterns before they reach your LLM model. It flags likely prompt-injection attempts so you can block, review, or route them differently—without making a model call.
