damngoodprompts

Recipes

Task-based starting points, not scaffolding to copy blindly. Each one is validated against specific models and rewritten from scratch against current guidance.

Agentic coding workflows

Brief a coding agent with the goal, boundaries, and success criteria, then let the model choose the investigation path.

Last reviewed: August 8, 2026

Brainstorming and ideation

Generate useful options by giving the model a real constraint set, not an empty request for ideas.

Last reviewed: August 10, 2026

Code review prompts

Ask for risks, regressions, and missing tests instead of a generic code summary.

Last reviewed: August 10, 2026

Customer support reply drafting

Draft replies that are grounded in policy, account facts, and the agent's actual authority.

Last reviewed: August 10, 2026

Debugging and iterating prompts

Fix a failing prompt by isolating the failure instead of adding more instructions.

Last reviewed: August 10, 2026

Decision-making and tradeoff analysis

Give a reasoning model the decision, constraints, and risk tolerance, then ask for a recommendation you can audit.

Last reviewed: August 8, 2026

Image and multimodal prompting

Prompt around what the model can see, what it should inspect, and what uncertainty should look like.

Last reviewed: August 10, 2026

Long-document summarization

Turn large documents into useful summaries by naming the decision, audience, and evidence rules.

Last reviewed: August 10, 2026

Meeting notes and transcript analysis

Extract decisions, owners, open questions, and follow-ups without pretending the transcript is cleaner than it is.

Last reviewed: August 10, 2026

Natural-language-to-data-visualization

Prompt for charts by naming the comparison, data shape, and decision the chart should support.

Last reviewed: August 10, 2026

Research and analysis

Separate retrieval from synthesis, and choose the model surface that matches the part of the job you are doing.

Last reviewed: August 25, 2026

Roleplay and simulation prompts

Use simulations for practice and evaluation without confusing a role with the scoring criteria.

Last reviewed: August 10, 2026

Structured data extraction

Use the provider's schema feature when one exists, and make null handling part of the contract.

Last reviewed: August 25, 2026

System prompt design for products

Design durable product instructions that define scope, authority, tool use, and handoff behavior.

Last reviewed: August 10, 2026

Tutoring and explaining a concept

Get an explanation calibrated to your real starting point instead of asking for generic simplification.

Last reviewed: August 8, 2026

Writing and content drafting

Get a useful first draft by giving the model audience, constraints, examples, and a clear definition of done.

Last reviewed: August 8, 2026