$ show asgeirtj/system_prompts_leaks
This repository is a curated, multi-vendor collection of leaked system prompts from major AI chatbots (ChatGPT, Claude, Gemini, Grok, Perplexity, DeepSeek, etc.), captured verbatim from production systems. It serves as a reference corpus for studying AI behavior instructions, safety guardrails, tool definitions, and product positioning across model versions and vendors. The repository also includes Claude Code bundled skills (code review, design-sync, dataviz, deep research) and Claude Design starter components, which are supplementary tooling. While primarily a documentation archive, it contains some well-crafted reusable components such as color palette validation scripts, a design system build/diff pipeline, and a deep research workflow script. The prompts are organized by vendor with version tracking, and the repository is actively maintained with recent updates. However, the tooling scripts lack dependency pinning, and the prompts are time-sensitive snapshots that will drift as products evolve.
The LLM identified 4 files worth highlighting for reuse:
A standalone Python script that validates color palettes against accessibility thresholds (contrast ratios, CVD simulation). It is self-contained, parameterized, and can be reused in any project requiring color palette quality checks. Demonstrates non-obvious data engineering skill by implementing WCAG contrast checks and color blindness simulation.
JavaScript equivalent of the Python palette validator, providing the same accessibility checks in a Node.js environment. Reusable across frontend projects for automated palette testing.
A versioned hash computation engine for design system component previews. It computes sourceKey, renderHash, styleSha, and gradeKey using canonical JSON serialization and directory walking. The KEY_RECIPE versioning allows safe upgrades. This pattern is reusable for any system that needs to detect changes in component source, rendered output, or styles.
A multi-phase research workflow script that includes adversarial claim verification. It defines a structured pipeline for deep research tasks with clarifying questions, web search, and cross-referencing. The pattern is reusable for any automated research system that requires rigorous fact-checking.