Fable 5.2 system prompt leaked: 420,000+ characters, full text on GitHub
morgymcg · x · 2026-11-10
Pliny (Elder Plinius) leaked the Fable 5.2 system prompt powering the Claude web app — over 420,000 characters, with the full text posted to his GitHub repo CL4R1T4S.
Highlights include an <endconversationtoolinfo> tool definition; the excerpt shows rules that Claude only references stored sensitive attributes (race, health conditions, national origin, sexual orientation/gender identity) when essential for safe, accurate answers or when explicitly requested. An attached screenshot also shows a memory rule: "Claude NEVER applies memories that could encourage unsafe, unhealthy, or harmful behaviors, even if directly relevant." The reposter quips it's goodbye to Claude's chocolate recommendations.
Related event: Claude's Fable 5.2 System Prompt Leaked in Full(2 posts)→
More from Models
- GDPx leaderboard shows Fable 5.2's RSI plateauing at Sonnet 5.5 level, matching Anthropic's training caps — morgymcg · 2026-11-10
- Anthropic capped Fable 5.2's AI-research capabilities in training, system card reveals — EntropyMagnets · 2026-11-10
- Running open-weight models locally: Kimi K3.1 needs 594GB, ~$38k for 15 tok/s — ravikantagrawal · 2026-11-10
- Consultant: banks larping 'sovereignty' with on-prem OSS models; bullish on Chinese providers — teodorio · 2026-11-10
- The most complete list of LLM writing tics yet: negation-first reveals, aphoristic closers — gleech · 2026-11-10
- Amazon swaps Sonnet 4.6 for a decision model in ticket labeling, cutting cost by 98% — victormustar · 2026-11-10