Steven Gonsalvez

Software Engineer

asgeirtj/system_prompts_leaks: Extracted system prompts from Anthropic - Claude Fable 5, Opus 4.8, Claude Code, Claude Design. OpenAI - ChatGPT GPT-5.6, Codex GPT-5.6, GPT-5.5. Google - Gemini 3.5 Fla

Why CEREBRO kept it

System prompt extraction from frontier models; reveals mechanics.

The text below is an automated extraction of the article at https://github.com/asgeirtj/system_prompts_leaks, stored verbatim in the public cerebro-vault repository. Copyright remains with the original publisher (github.com).

As seen in The Washington Post: See the hidden rules behind AI. Then use them to rewrite this article. (May 11, 2026) The purpose of this repo is to document the System Prompt instructions for all the AI chatbots out there - Claude, ChatGPT, Gemini etc. | What | Date | Link | |---|---|---| | Kimi K2.6 | July 14, 2026 | System prompt | | Perplexity Deep Research | July 14, 2026 | System prompt | | DeepSeek | July 14, 2026 | chat.deepseek.com | | GPT-5.6 | July 10, 2026 | ChatGPT (Sol, extra high) · Codex | | Claude Sonnet 5 | July 1, 2026 | System prompt | | Claude Design (Opus 4.8 — full

Backlinks

Appeared in 6 briefings

6 briefings re-surfaced this signal. A high number here is a deduplication weakness in the pipeline, not a popularity score.

Related

Shares tags: ai/llm-mechanics

Also from github.com