Debugging Prompt Patterns

Feed errors with full context and ask for a hypothesis before a fix.

Prompt EngineeringIntermediate

Why

GitHub's 2024 survey found 92% of US respondents use AI coding tools to generate test cases, and 73% of open source respondents use AI tools for coding or documentation. Pasting only the error line makes the model guess.

Giving the error + stack + relevant code + what you tried yields a real hypothesis and a targeted fix instead of a speculative patch.

When

Use whenever an error is non-obvious or the first fix attempt failed. Avoid over-engineering prompts for simple, well-defined tasks where a one-line instruction suffices.

How

1. Paste the full error + stack trace. 2. Include the relevant file(s) via @file. 3. State what you already tried (avoids repeated suggestions). 4. Ask: "what's the most likely cause? explain before fixing." 5. Approve the hypothesis, then ask for the minimal fix.

Example

Error: "TypeError: Cannot read 'id' of undefined at router.ts:42" @file src/api/router.ts @file src/data/books.ts Tried: checked books is non-empty. Ask: "most likely cause? then minimal fix." Hypothesis: "books[req.params.id] is undefined when id not found → guard with optional chaining + 404."

Common Problems

⚠Pasting only the error line: model guesses — include stack + code.
⚠JSON parse error on model output → wrap with a fenced ```json block and add "respond with valid JSON only" to the system prompt.
⚠Error: Token limit exceeded (400 Bad Request) → split the prompt, summarize prior context, or switch to a larger context model.

Best Practices

✓Always include the stack trace and the relevant file.
✓Demand a hypothesis before a fix to catch wrong root causes.
✓Measure the outcome with a concrete metric (time saved, defect rate, test coverage) so the practice can be justified and improved.

FAQ

What is Debugging Prompt Patterns?

Feed errors with full context and ask for a hypothesis before a fix. It helps developers quickly grasp the core value and use cases of Debugging Prompt Patterns.

How do I get started with Debugging Prompt Patterns?

To get started with Debugging Prompt Patterns, focus on these points first: Paste the full error + stack trace.

What are common mistakes with Debugging Prompt Patterns?

Common mistakes when using Debugging Prompt Patterns include: Pasting only the error line: model guesses — include stack + code.; JSON parse error on model output → wrap with a fenced ```json block and add "respond with valid JSON only" to the system prompt.; Error: Token limit exceeded (400 Bad Request) → split the prompt, summarize prior context, or switch to a larger context model..

What are the best practices for Debugging Prompt Patterns?

Best practices for Debugging Prompt Patterns include: Always include the stack trace and the relevant file.; Demand a hypothesis before a fix to catch wrong root causes.; Measure the outcome with a concrete metric (time saved, defect rate, test coverage) so the practice can be justified and improved..

How do I learn more about Debugging Prompt Patterns?

Start with the official docs, then search real-world examples on GitHub. Validate patterns in a small scratch repo before applying them to production code.