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 Flash, 3.1 Pro, Antigravity. xAI - Grok, Cursor, Copilot, VS Code, Perplexity, and more. Updated regularly.
asgeirtj/system_prompts_leaks is drawing steady momentum at +128.3 stars/day (97th percentile in the tracked cohort), for a momentum score of 37.7/100. Forks are climbing too, at +21.0/day.
Low breakout odds over the next 14 days, led by star velocity (97th pctl) and fork velocity (96th pctl). Confidence is high given the available history.
Reconstructed from 284 snapshots — the time-series GitHub’s API doesn’t expose.
~40 forks/day, 3.2× the trailing baseline.
~31 forks/day, 3.1× the trailing baseline.
~77 forks/day, 5.1× the trailing baseline.
Live score, refreshed every 6 hours. Links back to this page.
[](https://breakwave.vercel.app/repo/asgeirtj/system_prompts_leaks)Transparent heuristic · logged for model training
Percentile rank within the tracked cohort. The score self-calibrates — it answers “accelerating vs. everything else,” not raw size.
~93 forks/day, 7.0× the trailing baseline.
~52 forks/day, 9.1× the trailing baseline.
~38 forks/day, 5.3× the trailing baseline.
~24 forks/day, 3.1× the trailing baseline.
~42 forks/day, 3.8× the trailing baseline.
~32 forks/day, 4.2× the trailing baseline.
~18 forks/day, 3.4× the trailing baseline.
~14 forks/day, 4.7× the trailing baseline.
~17 forks/day, 3.6× the trailing baseline.