A curated list of LLM APIs that offer permanent free tiers for text inference, with base URLs, models, and rate limits.
A collection of security testing wordlists covering usernames, passwords, URLs, fuzzing payloads, web shells, and sensitive data patterns.