Results for “prompt-injection-defense”

Defensive review of an LLM/agent application for OWASP-LLM-style risks — prompt injection, unsafe tool calls, excessive agency, and PII/secret leakage. Reads your code, flags issues with file:line. Pure Claude Code, no signup.

Orchestrates a complete bulk RNA-seq differential-expression study from raw FASTQ reads through QC, alignment, quantification, differential expression, pathway enrichment, and publication figures.