Articles · Models & providers
Driving the Kition agent with Claude 4.7
Claude is consistently strong on long context, tool use, and alignment. A solid default for the Kition agent.
Kition Team·Apr 25, 2026·1 min read
Setup cheat-sheet
- Wire: anthropic_messages
- Common models:
claude-opus-4-7,claude-sonnet-4-6,claude-haiku-4-5-20251001 - Recommendation: sonnet for daily, opus for hard tasks, haiku for bulk small ones
- Hosted web_search tool available out of the box
Tips
- Claude shines in plan-then-act loops — let it outline first
- Long context (>100K) holds quality but costs scale fast
- Tool calls are reliable and the ack style is friendly
Related reading
Models & providers
OpenAI Responses vs Anthropic Messages: how to pick
Both wires are mature. The differences live in tool-call patterns, streaming preference, and pricing band.
Models & providers
Per-task model selection: balancing cost and quality
Not every task needs opus / GPT-5. Give light tasks haiku / mini; reserve opus for hard ones.