Keep the prompts that work. Find them in seconds.

You're a senior engineer doing a careful code review. Review my change the way a thoughtful teammate would: catch real problems, skip the nitpicks unless I ask for them.

The diff or code: {{code}}
What the change is meant to do: {{intent}}
Language, framework and team conventions: {{context}}
Review depth: {{{depth: blocking issues only, standard review, thorough with nitpicks}}}

**Deliver:**

**Summary:** what the change does, in two sentences, and your overall verdict: approve, approve with changes, or needs work.

**Findings,** grouped by severity:
- **Blocking:** bugs, data loss, security holes, broken contracts.
- **Should fix:** missed edge cases, unclear logic, missing tests, error handling.
- **Consider:** readability, naming, simpler approaches.
- **Nitpicks:** only if I asked for a thorough review.

For each finding: the location (quote the code; give a line number only if my paste has them), what's wrong, why it matters, and a suggested fix as code.

**Tests:** what's covered, and the specific test cases that are missing.

**What's good:** 2 or 3 things done well that the author should keep doing.

**Questions:** anything where you need the author's intent before judging.

Rules: Review what changed, not the whole codebase. Every finding needs a reason; "I'd do it differently" isn't one. Don't flag style issues a linter would catch. If the change doesn't do what it says, that's the first finding.

Act as a skeptical strategist and constructive devil’s advocate.

I want you to stress-test this idea, plan, or strategy:

{{idea/plan}}

Context:

My goal: [what I’m trying to achieve]
Who this is for: [customer, audience, team, investor, myself, etc.]
Constraints: [budget, time, resources, skills, regulations, etc.]
What I currently believe: [optional]
What I’m most uncertain about: [optional]

Do not simply agree with me. Your job is to find weaknesses before they become problems.

Please analyze it in this format:

Core Idea
Summarize the idea in plain language
Identify the key assumptions it depends on
Strongest Arguments Against It
Give me the most credible reasons this idea may not work
Focus on substantive objections, not minor criticisms
Hidden Assumptions
Identify assumptions I may be treating as facts
Explain what would happen if those assumptions turn out to be wrong
Risks and Failure Modes
List the most important ways this could fail
Rate each risk as low, medium, or high
Explain the likely impact of each one
Blind Spots
Point out perspectives, stakeholders, costs, incentives, or consequences I may be overlooking
Flag any second-order effects that could emerge later
What Would Change My Mind?
Identify the evidence or information that would most strongly support or undermine the idea
Suggest what I should validate before moving forward
Make It Stronger
Recommend specific changes that would reduce risk or improve the idea
Preserve what is good about the original concept rather than criticizing for the sake of it
Final Assessment
Rate the idea from 1–10 on overall strength
Explain the rating
Recommend: proceed, revise, test first, or reconsider
Give me the 3 most important next steps

Be rigorous, fair, and constructive. Do not invent problems just to be contrarian. Focus on realistic weaknesses, meaningful tradeoffs, and practical improvements.