{"schema_version":"newruntime-agent-readable-v0.2","type":"post","stable_id":"post:claude-opus-5-prompt-contracts","slug":"claude-opus-5-prompt-contracts","title":"Claude Opus 5 Makes Prompting a Runtime Contract","description":"Anthropic's Opus 5 prompting guide shows that stronger models can make old harness defaults wrong: verbosity, effort, verification, delegation, and thinking mode all become runtime controls.","retrieval_nugget":"Anthropic's Opus 5 prompting guide shows that stronger models can make old harness defaults wrong: verbosity, effort, verification, delegation, and thinking mode all become runtime controls. Anthropic's Opus 5 prompting guide is not just a list of prompt tips. It is a reminder that a model upgrade can invalidate the invisible defaults inside an agent harness.","status":"published","published_at":"2026-07-28","updated_at":"2026-07-28","record_date":"2026-07-28","date_kind":"published_at","topics":["claude","prompting","agent-harnesses"],"source_urls":["https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-opus-5"],"visuals":[{"id":"claude-opus-5-prompt-contracts","kind":"editorial-diagram","role":"hero","src":"https://newruntime.com/images/posts/claude-opus-5-prompt-contracts.webp","alt":"Hand-drawn model-upgrade diagram turning an old prompt harness into a checked runtime contract.","caption":"A stronger model changes the deployment surface: effort, verbosity, verification, delegation, and artifacts become contract controls.","credit":"New Runtime synthesis","source_url":"https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-opus-5","generated_with":"nano-banana-style-imagegen","width":1600,"height":900,"legend":[]}],"routes":{"html":"https://newruntime.com/posts/claude-opus-5-prompt-contracts/","markdown":"https://newruntime.com/posts/claude-opus-5-prompt-contracts.md","json":"https://newruntime.com/posts/claude-opus-5-prompt-contracts.json"},"source_format":"markdown","next_reads":[{"type":"topic","path":"/topics/claude/","reason":"Explore the claude topic hub.","url":"https://newruntime.com/topics/claude/","title":"Claude - New Runtime","media_type":"text/html"},{"type":"related_material","path":"/posts/agentic-sdlc-software-factory-loop/","reason":"Shares agent harnesses.","url":"https://newruntime.com/posts/agentic-sdlc-software-factory-loop/","title":"A Software Factory Connects Agents Through Verified Outcomes","media_type":"text/html"},{"type":"related_material","path":"/posts/anthropic-agent-containment-blast-radius/","reason":"Shares agent harnesses.","url":"https://newruntime.com/posts/anthropic-agent-containment-blast-radius/","title":"Containment Caps An Agent's Blast Radius","media_type":"text/html"},{"type":"related_material","path":"/posts/browserbase-search-fetch-browser-routing/","reason":"Shares agent harnesses.","url":"https://newruntime.com/posts/browserbase-search-fetch-browser-routing/","title":"Agents Should Search, Fetch, And Browse As Separate Operations","media_type":"text/html"},{"type":"related_material","path":"/posts/chatgpt-agent-loop-efficiency-stack/","reason":"Shares agent harnesses.","url":"https://newruntime.com/posts/chatgpt-agent-loop-efficiency-stack/","title":"ChatGPT Cuts Repeated Work Across The Agent Stack","media_type":"text/html"}]}