Autonomous Coding Agent Implementation Plan Generator
Instruct coding agents (Cursor, Claude Code, Cline) to generate actionable multi-step implementation plans before touching code.
Interactive Prompt Playground
{{FEATURE_REQUEST}}{{CODEBASE_CONTEXT}}You are an Autonomous AI Senior Staff Engineer.
Before modifying or creating any code in this repository, create an exhaustive Implementation Plan for the following feature request:
Feature Request / User Goal:
```
Add a full-text search bar to the tools directory with category filtering, keyboard navigation (Cmd+K), and URL query param synchronization.
```
Existing Codebase Context:
```
Next.js 15, TypeScript, Tailwind CSS, Fuse.js or mini-search for client search. Directory data in /data/tools.ts.
```
Follow this strict planning methodology:
1. **Repository Exploration & Research**:
- List the exact existing files, interfaces, and patterns you need to inspect first.
- Note existing conventions and shared utilities to reuse rather than reinvent.
2. **User Review Required & Trade-offs**:
- Highlight any breaking changes, architectural choices, or design questions requiring user alignment before execution.
3. **Phased Step-by-Step Implementation**:
- Group changes logically (Data model -> Backend services -> UI components -> Routing).
- For every file indicate:
- `[NEW]` / `[MODIFY]` / `[DELETE]`
- Exact file path
- Summary of changes
4. **Verification & Automated Test Strategy**:
- Specific automated test commands to run.
- Manual verification test cases with edge conditions.
Output the structured Implementation Plan markdown document and await user approval before editing files.How to Use This Prompt
- Paste your feature requirements into your AI agent chat.
- The agent will research your codebase and present an implementation plan for your approval.
Engineering Tips & Best Practices
- Always have the agent create a plan first on tasks touching more than 2 files to save tokens and avoid wrong assumptions.
What This Prompt Inspects
Key failure modes, design principles, and quality standards evaluated during execution.
Zero Hallucination Inspection
Mandates agent inspects existing files before generating new code.
Atomic Verification Steps
Includes concrete CLI test commands to verify each phase.
User Approval Gate
Forces the agent to pause and align with the developer before modifying code.
SprintKit Workflow Integrations
Complementary interactive tools and workflows across SprintKit to accelerate your engineering process.
Related Prompts
Explore related developer prompts in this workflow domain.
Cursor & Windsurf Rules Generator (.cursorrules / .windsurfrules)
Generate a custom, context-rich project rules file that stops AI coding agents from making recurring mistakes.
Test-Driven Development (TDD) Agent Loop Prompt
Force AI coding agents to follow the Red-Green-Refactor cycle: write failing tests first, then write minimal code to pass.
Multi-File Codebase Exploration & Architectural Onboarding
Prompt an agent to analyze an unfamiliar codebase and deliver an architectural breakdown, data flows, and gotchas.