//
Search across all documentation pages
9 pages in this section.
Learn why explicit, opinionated rules prevent Claude production incidents. Understand how ad hoc prompting leads to failures and how rules improve reliability.
Quickly reference core Claude rules for model selection, prompt hygiene, tool design, cost control, and security guardrails in production.
Match tasks to Claude models (Haiku, Sonnet, Opus) based on complexity, latency, and cost. Learn default tiers and when to escalate or de-escalate.
Design safe, composable Claude agents with these rules for scoping, naming, and validating tools. Prevent hidden risks as toolsets grow.
Tech leads: Enforce these security guardrails for Claude in production. Learn rules for scoping permissions, validating output, and logging tool calls.
Learn best practices for using Claude effectively in production. This checklist covers model selection, production readiness, and team audits.
A single-page roundup of every highlight bullet from the 8 pages in the Claude Rules section, grouped by source page so you can scan all 25 takeaways without opening each article individually.