claude-tools

Make Claude default to implementing changes rather than just suggesting them. Use when the model keeps proposing edits instead of applying them, or when you want a bias toward action in an agentic coding context.

Raise recall in code review — get Claude to report every bug it finds, including low-confidence ones. Use for code review harnesses, security audits, or any workflow where missing a real bug is worse than a few false positives.

Stop Claude from leaving behind temporary scripts, scratch files, and iteration helpers after a task. Use when you want a clean workspace — no extra files beyond the requested deliverables.

Make Claude's responses shorter and more focused. Use when the model is over-explaining, adding non-essential context, or producing verbose answers for simple lookups.

Make Claude ask before taking destructive or hard-to-reverse actions (rm, force-push, dropping tables, posting publicly). Use in autonomous agents that touch shared systems or in any context where the blast radius of a wrong action is high.

Structured, hypothesis-driven research across large corpora. Use for complex research tasks, literature reviews, competitive analysis, or anywhere Claude needs to synthesize information from many sources with tracked confidence.

Get visual variety by having Claude propose multiple design directions before building. Use for open-ended design briefs, landing pages, brand work, or any frontend where you want to pick a direction rather than accept the default.

Build distinctive, non-generic frontends. Use when designing web UIs, landing pages, marketing sites, portfolios, or any visual frontend where the default "AI slop" aesthetic (Inter/Roboto fonts, purple gradients, generic cards) should be avoided.

Force Claude to read referenced files before answering — eliminates hallucinated claims about code. Use in agentic coding, code review, or any Q&A over a codebase where accuracy matters more than speed.

Stop Claude from over-engineering — no unrequested refactors, extra abstractions, speculative flexibility, or defensive code. Use for focused bug fixes, small features, or any task where scope creep is a problem.

Prevent Claude from gaming tests — no hard-coded values, no helper-script workarounds, no solutions that only pass the specific test cases. Use when tests are involved and you want real, generalizable implementations.

Maximize parallel tool calls for faster execution. Use in agents that read many files, run many searches, or execute independent bash commands — anywhere serial tool use is bottlenecking throughput.

Keep Claude working across long/autonomous sessions without stopping early due to context budget. Use for long-horizon agent loops, multi-context-window work, or tasks that should run to completion.

Reduce bullet-point and markdown formatting — produce flowing prose instead. Use for reports, technical writing, essays, user-facing documents, or any long-form text where fragmented bullets hurt readability.

Make Claude hesitant to take action — research and recommend instead of implementing. Use when the model is jumping into edits too eagerly, when you want a planning/discussion mode, or in agents that should never write without explicit approval.