How to prompt each major AI model, from the team behind Anvyll.https://anvyll.apphttps://anvyll.app/learn/claude-opus-5-prompting-guidehttps://anvyll.app/learn/claude-opus-5-prompting-guideA practical Claude Opus 5 prompting guide: XML-tagged structure, controlling verbosity and scope, effort levels, and what to remove from older Claude prompts.Fri, 11 Sep 2026 00:00:00 GMTAnthropicClaude Opus 5https://anvyll.app/learn/claude-sonnet-4-5-prompting-guidehttps://anvyll.app/learn/claude-sonnet-4-5-prompting-guideClaude Sonnet 4.5 prompting guide: XML-tagged structure, stating the full task at once, bounding scope, and one skeleton for code, writing and analysis.Fri, 11 Sep 2026 00:00:00 GMTAnthropicClaude Sonnet 4.5https://anvyll.app/learn/claude-haiku-4-5-prompting-guidehttps://anvyll.app/learn/claude-haiku-4-5-prompting-guideClaude Haiku 4.5 prompting guide: short XML-tagged briefs, one deliverable per prompt, numeric length caps, and why Opus-style prompts backfire on it.Fri, 11 Sep 2026 00:00:00 GMTAnthropicClaude Haiku 4.5https://anvyll.app/learn/gpt-6-astra-prompting-guidehttps://anvyll.app/learn/gpt-6-astra-prompting-guideGPT-6 Astra prompting guide: outcome-first markdown sections, getting it to act instead of ask, instruction priority, and taming its heavy formatting.Fri, 11 Sep 2026 00:00:00 GMTOpenAIGPT-6 Astrahttps://anvyll.app/learn/gemini-3-pro-prompting-guidehttps://anvyll.app/learn/gemini-3-pro-prompting-guideGemini 3 Pro prompting guide: context first and constraints last, positive grounding instead of "do not infer", split-step verification and personas.Fri, 11 Sep 2026 00:00:00 GMTGoogleGemini 3 Prohttps://anvyll.app/learn/gemini-3-flash-prompting-guidehttps://anvyll.app/learn/gemini-3-flash-prompting-guideGemini 3 Flash prompting guide: dense labeled blocks, constraints as the final block, positive grounding, and the long-brief habits that hurt it.Fri, 11 Sep 2026 00:00:00 GMTGoogleGemini 3 Flashhttps://anvyll.app/learn/grok-4-6-prompting-guidehttps://anvyll.app/learn/grok-4-6-prompting-guideGrok 4.6 prompting guide: a GOAL / CONTEXT / OUTPUT FORMAT / QUALITY BAR skeleton, static content first, reasoning effort, and an uncertainty rule.Fri, 11 Sep 2026 00:00:00 GMTxAIGrok 4.6https://anvyll.app/learn/composer-2-5-prompting-guidehttps://anvyll.app/learn/composer-2-5-prompting-guideCursor Composer 2.5 prompting guide: a goal-first coding brief with repo context, up-front constraints, diff-only output and precise follow-ups.Fri, 11 Sep 2026 00:00:00 GMTCursorComposer 2.5https://anvyll.app/learn/llama-4-prompting-guidehttps://anvyll.app/learn/llama-4-prompting-guideLlama 4 prompting guide: outcome-first markdown headers, removing leftover system prompts and personas, explicit rule priority, and a clear output contract.Fri, 11 Sep 2026 00:00:00 GMTMetaLlama 4https://anvyll.app/learn/mistral-large-prompting-guidehttps://anvyll.app/learn/mistral-large-prompting-guideMistral Large prompting guide: compact system-style labeled blocks, strong action verbs, constraints as bullets, and the chatty phrasing that weakens results.Fri, 11 Sep 2026 00:00:00 GMTMistral AIMistral Largehttps://anvyll.app/learn/perplexity-sonar-prompting-guidehttps://anvyll.app/learn/perplexity-sonar-prompting-guidePerplexity Sonar prompting guide: a mandatory citation policy, no invented URLs, findings kept apart from interpretation, and tightly scoped searches.Fri, 11 Sep 2026 00:00:00 GMTPerplexityPerplexity Sonar