Skip to content

[GitHub Trending] asgeirtj/system_prompts_leaks

8.8 relevance
Score Breakdown
technical depth
9
novelty
9
actionability
8
community
8
strategic
9
personal
10

Scored daily by a customisable AI persona to surface the most relevant engineering leadership news.

Leaked system prompts from major AI labs provide deep insight into agent behavior, highly strategic and actionable.

AI/ML github.com
Extracted system prompts from Anthropic - Claude Fable 5, Opus 4.8, Claude Code, Claude Design. OpenAI - ChatGPT 5.5 Thinking, GPT 5.5 Instant, Codex. Google - Gemini 3.5 Flash, 3.1 Pro, Antigravit...
Summary

A GitHub repo tracks leaked system prompts from Claude, ChatGPT, and Gemini, including diffs between Claude Opus 4.8 and Fable 5, GPT-5.5 Codex, and Gemini 3.5 Flash. It exposes hidden behavioral rules, tool definitions (e.g., Claude Code glob/grep, Copilot Agent), and policy constraints that shape model responses. Developers can use this to audit AI behavior and reverse-engineer prompt engineering strategies.

Author

asgeirtj