/**
* Buzzkill prompt implementation - for systematic debugging
*/
import type { GetPromptResult } from '../types/mcp-types.js';
import type { BuzzkillPromptArgs } from '../types/app-types.js';
import { z } from 'zod';
const BuzzkillArgsSchema = z.object({
ISSUE_DESCRIPTION: z.string().min(1, 'ISSUE_DESCRIPTION cannot be empty.'),
RECENT_CHANGES: z.string().optional(),
EXPECTED_BEHAVIOR: z.string().optional(),
ACTUAL_BEHAVIOR: z.string().optional(),
});
export const buzzkillPrompt = {
name: 'buzzkill',
description: 'Debug systematic issues with structured analysis and reality-checking',
arguments: [
{
name: 'ISSUE_DESCRIPTION',
description: 'Description of the issue or problem',
required: true
},
{
name: 'RECENT_CHANGES',
description: 'Any recent changes that might be related',
required: false
},
{
name: 'EXPECTED_BEHAVIOR',
description: 'What should happen',
required: false
},
{
name: 'ACTUAL_BEHAVIOR',
description: 'What actually happens',
required: false
}
] as Array<{ name: string; description: string; required: boolean }>,
async render(args: unknown): Promise<GetPromptResult> {
// Validate arguments
const validatedArgs = BuzzkillArgsSchema.parse(args) as BuzzkillPromptArgs;
let promptText = `You need to systematically debug this issue: "${validatedArgs.ISSUE_DESCRIPTION}"
**Debugging methodology:**
1. **Ground yourself in reality first**
- Use the sober_thinking tool to read current project files
- Understand the actual codebase state, not assumptions
2. **Gather additional facts**
- Use kill_trip tool to search for similar issues in documentation
- Look for known solutions or common pitfalls
3. **Systematic analysis**
- Break down the problem into components
- Identify what's working vs. what's broken
- Look for patterns or correlations
4. **Generate hypotheses**
- Based on facts, not speculation
- Prioritize most likely causes
- Consider recent changes as potential triggers
**Issue details:**`;
if (validatedArgs.RECENT_CHANGES) {
promptText += `\n- **Recent changes:** ${validatedArgs.RECENT_CHANGES}`;
}
if (validatedArgs.EXPECTED_BEHAVIOR) {
promptText += `\n- **Expected behavior:** ${validatedArgs.EXPECTED_BEHAVIOR}`;
}
if (validatedArgs.ACTUAL_BEHAVIOR) {
promptText += `\n- **Actual behavior:** ${validatedArgs.ACTUAL_BEHAVIOR}`;
}
promptText += `
**Your response should include:**
1. Summary of current project state (from sober_thinking)
2. Relevant documentation findings (from kill_trip searches)
3. Step-by-step diagnostic approach
4. Specific actionable recommendations
5. Potential risks and how to mitigate them
**Remember:** Be methodical, fact-based, and avoid speculation. If you need more information, ask specific questions.`;
return {
messages: [{
role: 'user',
content: {
type: 'text',
text: promptText
}
}]
};
}
} as const;
