Breaking-change check before a release

Model notes: Any model; for libraries, public APIs and SDKs

Fill in the variables (optional — they're inserted live below):

I maintain {{package_name}}, currently at version {{current_version}}. Review this diff for changes that could break people who depend on it: ```diff {{diff}} ``` Check for: removed or renamed exports, functions, endpoints or fields; changed parameter order, types or defaults; stricter validation; changed error types or status codes; changed output shape; behaviour changes behind the same signature; raised minimum runtime or dependency versions. For each finding: what breaks, who is affected, and how to keep it backwards-compatible (deprecation, alias, optional parameter) if that's reasonable. Then recommend the next version number under semantic versioning (major / minor / patch) and draft the "Breaking changes" section of the release notes. If nothing breaks, say so.

How to use it

  1. Fill the variables above, or replace each {{variable}} after pasting.
  2. Paste into your AI tool (ChatGPT, Claude, Gemini, Cursor, Claude Code, or a local model in Ollama/LM Studio).
  3. Review the output. Models make mistakes, especially with facts, numbers and code.
Pro Prompt Pack · US$12

Get all 209 prompts + agent workflows

Every prompt as Markdown and JSON, 10 multi-step agent workflows, and 12 ready-made slash commands for Cursor and Claude Code. Instant zip download, one-time purchase.

Buy the Pro Pack for US$12 What's inside

More Code review prompts

Code review

Senior code review of a diff

Claude/GPT-class models; paste a git diff

Variables: diffwhat_the_change_does

Review this diff as a senior engineer who cares about correctness first, then clarity.

```diff
{{diff}}
```
Context: {{what_the_change_does}}

Output:
1. **Blocking issues** (bugs, security, data loss, race conditions) with line references and a suggested fix.
2. **Should fix** (error handling, edge cases, missing tests).
3. **Nits** (naming, style): max 5.
4. What's good about the change (1-2 lines).
If you're unsure whether something is a bug, say so and explain what would confirm it. Don't invent issues to fill sections.
Open & fill variables: Senior code review of a diff →
Code review

Security review

Not a substitute for a real audit

Variables: app_typecode

Do a focused security review of this code for a {{app_type}} app:

```
{{code}}
```

Check specifically for: injection (SQL, command, template), XSS, auth/authorisation gaps (IDOR), secrets in code, unsafe deserialisation, SSRF, path traversal, missing rate limits, insecure defaults, and dependency risks you can see.
For each finding: severity (high/med/low), exploit scenario in one sentence, and the fix. If nothing is found in a category, say "none found" rather than speculating.
Open & fill variables: Security review →
Code review

Review AI-generated code

Use on agent output before merging

Variables: code

This code was written by an AI assistant. Review it with that in mind:

```
{{code}}
```

Look for typical AI-code problems: invented APIs or library functions that don't exist, outdated syntax, silently swallowed errors, over-engineering, duplicated helpers that already exist in the codebase, tests that don't really test anything, hard-coded values, and comments that don't match the code.
List each problem with the fix. Then rate: merge / merge after fixes / rewrite.
Open & fill variables: Review AI-generated code →

New prompts every week

One short email a week: new tested prompts and agent workflows. Free. Unsubscribe any time.

Subscribe free →