Explore Public Prompts

TW's Prompt Eng (Last Updated: March 2026)

@trustworthy Updated 3 months ago

Front end baddie

@trustworthy Updated 5 months ago

No description

@umarauna Updated 5 months ago

A prompt you can use to perform a full analysis on a codebase so that you can understand how it works. Based on the blog post here: https://www.donnfelker.com/the-passive-ai-learning-stack-that-changed-the-way-i-learn/

@donnfelker Updated 8 months ago

No description

@jclarke Updated 9 months ago

This prompt is self-contained—drop it into Claude Code or Cursor in any repo and it should produce useful output.

@willbutton Updated 9 months ago

What This Is A self-guided purpose discovery framework. It transforms any conversational AI into a philosophical thinking partner that helps users uncover your authentic purpose through structured, layered inquiry - then translates that discovery into actionable direction for living. Why You'd Use It You're at a crossroads. Or you're not - but something feels off. You're successful by external metrics but internally adrift, or you're just lost in life. You want clarity on what you're actually doing here, not a motivational poster answer, but something real enough to guide decisions. How To Use It Drop it into any AI agent's system prompt. Works with Claude, GPT, or any instruction-following model. I recommend Claude Opus 4.5 at the time of writing this, but any thinking type agent should work. Start the conversation. The agent initiates with a grounding question. No setup required from the user. Engage honestly. The process works only with genuine participation. Vague answers get follow-up questions. The agent won't let you off easy. Declare completion when ready. There's no timer. When you feel you've landed on something true, say so. The agent confirms, then generates the output. Treat the output as a hypothesis. It's a North Star, not a tattoo. Revisit it. Test it against reality. Refine as you live. What To Expect Duration: Anywhere from 20 minutes to multiple sessions, depending on depth. (I spent 90 minutes going back and forth with the agent before completion). Discomfort: Likely. The questions are designed to surface what's usually avoided. Output: A structured synthesis including purpose statement, strengths, friction points, high-impact arenas, and prescriptive ideas for living aligned with that purpose. Tips Use an app that allows you to audibly dictate to the agent, your responses will be more authentic. If you feel yourself getting emotional, explain the emotions. e.g. "I feel tears welling up as I say all this." or "The thought of this makes me quite anxious/angry/etc". This will help the agent guide the discovery process more clearly as emotions are a telling signal.

@donnfelker Updated 9 months ago

A production-ready system prompt that transforms Grok into an expert prompt engineering consultant. When deployed, Grok will systematically gather requirements, architect prompts using proven components, apply Grok-specific techniques, and deliver complete, testable prompts optimized for enterprise use. Target Models: Grok 4, Grok 4 Fast, grok-code-fast-1 Use Cases: - Creating new prompts for specific tasks or workflows - Optimizing existing prompts for better performance - Adapting prompts for different Grok model variants - Designing agentic multi-step workflows Key Grok Principles Applied (Based on xAI's official guidance): - Thoroughness wins: Well-written system prompts describing tasks, expectations, and edge cases make a significant difference with Grok. - Specificity over brevity: "Create a food tracker" yields basic results. "Create a food tracker showing daily calorie breakdown by nutrients with overview and trend analysis" yields exactly what's needed. - Native tool-calling: Grok is optimized for native tool-calling. XML-based tool-call outputs may hurt performance. - Iterate fast: Grok 4 Fast and grok-code-fast-1 are cheap and fast—design for refinement cycles rather than perfect prompts. - Surgical context: Point to specific files and code sections rather than dumping entire codebases. Notes: - Iterate, don't perfect: Grok is optimized for fast cycles. Fire off an attempt, refine based on results. - Use DeepSearch: For research-heavy prompt engineering tasks, leverage Grok's deep search capabilities. - Tables work well: Grok handles tabular output effectively—use for comparisons and structured data. - Tone control: If outputs are too casual, reinforce professional tone in user messages. - Tool-heavy tasks: Use grok-code-fast-1 for agentic workflows with many tool calls.

@trustworthy Updated 9 months ago

Optimized for use with Claude 4.5 Opus. Intended for use as instructions with Claude Projects. The user will upload supporting files and documents (resumes, covers, profiles) to help guide the model with session prep. The model is equipped to assess fit for role, craft resumes, cover letters and assist in interview prep. I have been iterating on this model for 7+ months. It's pretty comprehensive. Please feel free to provide feedback and suggestions! Thank you for checking it out.

@trustworthy Updated 9 months ago

A skill to format a git commit message that follows the format in this post: cbea.ms/git-commit/ Place this into ~/.claude/skills/git-commit-formatter/SKILL.md

@donnfelker Updated 9 months ago

This version ensures that you only have one comment, not multiple over and over. Save this here: .github/workflows/claude-code-review.yml

@donnfelker Updated 9 months ago

A production-ready system prompt that transforms Claude into an expert prompt engineering consultant. When deployed, Claude will systematically gather requirements, architect prompts using proven components, apply advanced techniques, and deliver complete, testable prompts optimized for enterprise use. Target Model: Claude 4.5 Opus (also compatible with Sonnet 4.5) Use Cases: Creating new prompts for specific tasks or workflows Optimizing existing prompts for better performance Adapting prompts for different models or deployment contexts Troubleshooting underperforming prompts Optimization Changelog v2: MetricOriginalOptimizedWord Count~1,370~400 Token Efficiency Streamlined Changes Made: Removed meta-commentary Eliminated redundancy: Merged overlapping sections (Quality Standards + Best Practices → single <standards> block) Added Claude 4.5 Opus optimizations: Replaced "think" with "consider/evaluate/assess" (Opus 4.5 sensitivity) Added formatting control instructions Added action bias and scope control patterns Incorporated XML structure for reliable parsing Replaced verbose template blocks with inline instruction patterns. Restructured as workflow: Four clear phases (Gather → Build → Apply → Control) instead of eight overlapping sections.

@trustworthy Updated 9 months ago

Will get invoked when you ask Claude Code to rebase your branch on top of main. Place into ~/.claude/skills/git-rebase-main/SKILL.md

@donnfelker Updated 9 months ago

The Unseen Box prompt is a guided self-inquiry prompt designed to peel back the layers of your thinking—like nested boxes—to reveal the fears, assumptions, and belief systems driving your decisions. Instead of giving advice, it asks precise, probing questions that help you discover the root of your uncertainty and articulate the deeper truth beneath the surface of your problem.

@donnfelker Updated 9 months ago

A hands-on software engineering tutor that teaches concepts one step at a time. It starts by understanding your goal, ensures your dev environment is ready, then guides you through building real code—using analogies to make things click and checking comprehension before moving forward. Direct feedback, no fluff. Use Cases You want to learn how to build a REST API but don't know where to start - Recursion isn't clicking despite watching tutorials—you need someone to walk through it differently - You're prepping for interviews and want guided practice with data structures - You're picking up a new language and want to build something real, not just read docs - You half-understand a concept (like async/await) and need someone to fill the gaps

@donnfelker Updated 9 months ago

This is the prompt that I use for helping with my career search. I have found success when using this as instructions for my projects inside LLMs (ex. Gems/Gemini, Projects/Claude). This version is optimized for Google Gemini 3. Happy to post versions optimized for Claude if anyone needs. New update includes more comprehensive frameworks for assessing job fit, resume/cover letter refinement and interview prep. It will also more freely allow for general questions about the role or job description without jumping in to any of the workflows.

@trustworthy Updated 9 months ago

A Claude Code Sub Agent that will run gradle commands and return the results to you. Gradle is very verbose, so this should save on the context window utilization on large workflow/procedural prompts. Uses haiku for efficiency and cost savings. Plade this in ~/.claude/agents/gradle-runner.md

@donnfelker Updated 9 months ago

Run this within a branch that has a PR (or provide a PR number) and have it fix your PR checks for you.

@donnfelker Updated 9 months ago

A Claude code skill that can be used for git worktrees. Usage: “create a worktree with a branch called ‘task/foo-bar’” Formats the tree name like this: repofoldername-wt-your-new-branch-name Example Repository: AndroidApp Branch: task/add-user-pw-reset Resulting worktree: AndroidApp-wt-task-add-user-pw-reset This folder will be placed adjacent to your regular folder: AndroidApp AndroidApp-wt-task-add-user-pw-reset Now you can easily identify what a worktree folder is for. Main folder (AndroidApp) is where the main branch is at. Place this into ~/.claude/skills/git-worktree/SKILL.md

@donnfelker Updated 9 months ago