Claude Sonnet 5.5 Prompt Generator

Triage these emails. For each one, give me: a one-line summary, the action I need to take (reply / delegate / archive / schedule), and a draft reply under 60 words if a reply is needed. Put anything with a deadline in the next 48 hours at the top. [Paste emails]

Write a commit message and a pull request description for this diff. Commit message: imperative mood, subject line under 60 characters, then a short body explaining why the change was made. PR description: Summary (2–3 bullets), How to test (numbered steps), Risks (anything a reviewer should look at closely). [Paste git diff]

Turn these meeting notes into a table with columns: Action, Owner, Due date, Status. If an owner or date was not stated, write "unassigned" or "no date" instead of guessing. Below the table, list any decision that was made and any question left open. [Paste notes or transcript]

Write a [PostgreSQL / MySQL / BigQuery] query for this request: [describe what you want, e.g. 'monthly active users for the last 6 months, split by plan']. Schema: [Paste table definitions] Return only the query, then one sentence explaining any assumption you made about the data.

Review this pull request like a senior engineer on my team. Sort findings into: Bugs (will break something), Risks (could break under some condition), and Suggestions (optional). For each finding give the file and line, what is wrong, and the smallest fix. Do not comment on formatting a linter would catch. Context: [what the PR is supposed to do] [Paste diff]

Turn my rough notes into a 1,200-word blog post for [audience]. Keep my opinions and examples; cut repetition. Structure: a hook that states the problem in the first two sentences, 4 sections with descriptive H2s, and a short conclusion with one concrete next step. Write in plain, direct English with no filler phrases. [Paste notes]

Analyse this data and answer: [your question, e.g. 'which product lines are growing and which are shrinking?']. Give me: the answer in 3 sentences, the 3 numbers that support it, one chart I should build (type, x-axis, y-axis), and any data quality problem that could change the answer. [Paste CSV or table]

Write 8 customer support reply templates for [your product] covering: refund request, bug report, feature request, billing error, cancellation, login problem, slow response complaint, and a thank-you. Each template: under 100 words, warm but brief, with [placeholders] for names and details, and one line telling the agent when not to use it.

Find the cause of this error. Error and stack trace: ``` [Paste] ``` Relevant code: ``` [Paste] ``` What I expected: [expected behavior] What I tried: [attempts so far] Give me: the most likely cause and the evidence for it, the fix as a code change, and a test that would have caught it. If two causes are equally likely, tell me how to tell them apart.

Rebuild the interface in this screenshot as a [React + Tailwind / plain HTML + CSS] component. Match the layout, spacing and hierarchy closely; use semantic HTML and accessible labels. Replace real text with the same text, and images with neutral placeholders. After the code, list anything in the screenshot you could not read clearly and what you assumed instead. [Attach screenshot]

Compare the implemented page (screenshot 1) with the design (screenshot 2). List every visible difference: spacing, alignment, font size or weight, color, missing or extra elements, and text changes. For each one, rate it Must fix / Should fix / Minor and say which element it affects. End with the 3 fixes that close most of the gap. [Attach both screenshots]

Write a test suite for this module using [Jest / pytest / Go test / your framework]. Cover: the normal path, edge cases (empty input, very large input, invalid types), and error handling. Name each test after the behavior it checks. After the tests, list any behavior that looks like a bug rather than a design choice. [Paste module]

Review this API design before we build it. Check: naming consistency, HTTP methods and status codes, pagination, error format, versioning, and anything that will be hard to change after clients depend on it. Propose a corrected version of any endpoint you would change, and explain why in one sentence each. [Paste OpenAPI spec or endpoint list]

You have access to a terminal in this repository. Goal: get the test suite passing without disabling or skipping any test. 1. Install dependencies and run the tests. Report what fails. 2. For each failure, find the root cause before changing code. 3. Make the smallest change that fixes it, then re-run the tests. 4. Stop and ask me if a fix would change public behavior. When done, summarise every change with the file, the reason and the test it fixed.

Implement this feature: [describe the feature and who uses it]. First, write a short plan: files to change, new files, data model changes, and risks. Wait for my OK. Then implement it in small steps, each one leaving the app working. Include tests for the new behavior and update any documentation the change makes wrong. Relevant files: [Paste or point to files]

Upgrade this project from [framework/library + current version] to [target version]. Read the official migration notes I paste below, then: list every breaking change that affects this codebase with the file it touches, apply the changes one area at a time, and flag anything that needs a manual decision. Migration notes: [Paste] Codebase: [Paste key files or give repository access]

Read these sources and write a 1-page brief answering: [your question]. Structure: the answer in 3 sentences; the key evidence with the source each point comes from; where the sources disagree and which one is more reliable and why; and what is still unknown. Use only the sources provided and say so when they do not answer something. [Paste or attach sources]

Build a working [describe the app, e.g. 'habit tracker with login, daily check-ins and a weekly chart'] using [stack]. Work in checkpoints: 1) project setup and a running empty app, 2) data model and storage, 3) core feature, 4) UI polish, 5) tests and a README. After each checkpoint, run the app and tests, report what works, and list anything you chose that I might want to change. Do not move to the next checkpoint with failing tests.

Review the authentication and session handling in this code for security problems: token storage, expiry and refresh, password reset, rate limiting, CSRF, session fixation, and anything that trusts user input. For each issue: severity (Critical / High / Medium / Low), how it could be exploited in one sentence, and the fix as a code change. List what you checked and found no problem with, too. [Paste auth-related code]

This [endpoint / page / job] is slow: [describe symptoms, e.g. 'p95 of 4s under normal load']. Using the code, query plans and profiler output below, find where the time goes. Rank the causes by how much time each costs, propose a fix for the top 3 with the expected improvement, and tell me what to measure to confirm each fix worked. [Paste code, EXPLAIN output, profiler traces]