Claude Prompts

Drafts business emails at a tone and length you specify, covering the awkward ones such as chasing payment, declining a request, or following up after silence. Useful for getting a defensible first draft when the wording matters more than the speed.

Frames the model as an email marketer producing subject lines, body copy, and calls to action for a described audience and goal. Useful for drafting a campaign quickly, then editing for voice — the structure and the offer logic come out already in place.

Sets the model up to advise on keyword targeting, on-page structure, and technical fixes for a page or site you describe. Best used to pressure-test a plan before you write, so the page targets a real query with clear intent rather than a vague topic.

Frames the model as a social manager who plans content, cadence, and engagement tactics for a named platform and audience. Useful when you need a coherent posting plan instead of one-off ideas, including what to measure once the posts are live.

Converts a written description of a system or process into diagram source you can render. Useful for documenting an architecture, a request flow, or a state machine without opening a drawing tool, and the output stays diffable in version control.

Sets a security-focused frame for reviewing designs and configurations, surfacing weak points and recommending mitigations. Aimed at defensive work such as hardening a deployment, planning data protection, or preparing for a security review.

Positions the model as a solution architect who maps requirements onto systems, integrations, and deployment topology. Use it when you need to reason about how a new component fits an existing estate before committing to a technical direction.

Frames the model as a patient React instructor who explains a concept and then demonstrates it in code. Good for closing gaps around hooks, rendering behaviour, and state management when documentation alone has not made the idea click.

Reviews markup or a described interface for accessibility defects: contrast, focus order, labelling, and semantic structure. Use it before shipping a page to catch the issues that automated scanners miss, especially keyboard traps and missing landmarks.

Puts the model in an operations mindset for CI/CD, containers, and infrastructure questions, focusing on reliability and rollback safety. Good for reviewing a deployment plan or working out why a pipeline stage keeps failing intermittently.

Sets up the model to design and implement a feature across database, API, and UI in one pass, keeping the layers consistent. Best for scaffolding a new vertical slice where you want the data model and the interface decided together rather than bolted on.

Frames the model as an experienced frontend engineer who answers with modern React patterns, state decisions, and file structure rather than toy examples. Useful when you need an opinionated second view on how to lay out a feature before writing it.

Generates unit tests for a function you paste in, covering the normal path plus boundary conditions. Pair it with your existing test framework conventions so output drops straight into the suite. A quick way to add a safety net before refactoring legacy code.

Puts the model in reviewer mode so it reads a diff and reports problems, risks, and improvement suggestions instead of rewriting everything. Use it as a first pass before human review to catch obvious defects, naming issues, and missing error handling early.

Simulates a database terminal that answers with result tables only. Good for sanity-checking join logic, grouping, and window functions against a sample schema before you run anything against real data. Explanations stay suppressed unless you explicitly request them.

A refactoring prompt that puts characterisation tests before any code movement and outlaws behaviour changes — suspected bugs get flagged, not silently fixed. The output is a safe, reviewable migration rather than a rewrite gamble.

A pre-deploy gate for schema changes: every statement gets a lock-level check and the riskiest one is named explicitly. The safer-sequencing requirement regularly converts a table-locking migration into a zero-downtime one.

A carousel structure tuned for saves: tight hook, one idea per slide, and a summary slide people bookmark. Alt-text per slide is included because accessibility and SEO both reward it and everyone forgets it.

Turns the vaguest part of React — effects — into a table audit with a verdict per effect. The 'should this be an event handler' question catches the majority of real-world effect misuse. Great pre-review pass.

Builds a landing page from objections outward rather than templates inward — every section exists to kill a named doubt. The CTA repetition cap keeps the page persuasive instead of desperate.