Description

A comprehensive research collection archiving leaked system prompts and internal tool definitions for frontier AI models. The repository features a 950,000-character deep dive into the Claude Fable 5.2 system prompt, documenting its complex safety guardrails, tone definitions, and the 'antml' function-calling framework used by Anthropic's coding agent.

Descriptions, tags, and model credits may be AI-generated or inferred from public sources and can be incomplete or wrong. Learn more.