Provides prompting templates and phased interaction strategies for AI-assisted development workflows.
npx skills add https://github.com/jasonkneen/kiro --skill ai-prompting
Please help me install this Agent Skill. Skill: ai-prompting Source: https://github.com/jasonkneen/kiro/tree/main/skills/ai-prompting Command: npx skills add https://github.com/jasonkneen/kiro --skill ai-prompting
Effective prompting strategies for AI-assisted development, enabling clearer context, phased collaboration, and reliable results.
Provide a context-rich prompt that starts an AI-assisted coding session using context-first prompting.
High-intent search queries and answers about installing and using this skill.
Phased interaction breaks complex coding tasks into sequential steps for an AI assistant. By using iterative refinement, you progressively review and adjust outputs, which significantly improves the overall quality and accuracy of the final code.
Role-based prompting assigns a specific expert persona to the AI, while validation prompts ask it to check its own logic against requirements. Applying these techniques in product development ensures higher quality outputs and reliable task execution.
Yes, these prompting strategies apply to any AI-assisted workflow. The templates and examples support design tasks and general product development by structuring clear context, phased collaboration, and iterative refinement for the AI.
Irrelevant code generation often occurs when prompts lack sufficient context. By applying context-first prompting templates, you provide the AI with necessary project background immediately, ensuring the generated code aligns with your actual goals.
The best way to structure prompts for iterative refinement is using a phased interaction approach. Start with a context-rich prompt outlining the goal, then sequentially request modifications and validate outputs to progressively improve the results.
HuijiroGather design context and persist it to a config file.
ShaheerKhawajaRefine flagged outputs through structured critique and iterative improvement passes.
majiayu000Quantify AI-generated code quality using pass@k and pass^k metrics.
ReinaMacCredyTransform unstructured user prompts into detailed instructions for AI agents.
