
Claude Code subagents: status updates cost 99% less
Claude Code subagents now write status lines with Haiku 5.5, cutting per-update cost about 99%. Session savings are closer to 5%. Here is who gains.
Tag archive

Claude Code subagents now write status lines with Haiku 5.5, cutting per-update cost about 99%. Session savings are closer to 5%. Here is who gains.

Claude Code subagent allowed-tools explained: the real field is tools, plus disallowedTools, MCP wildcards, inherit vs restrict, and hooks that enforce it.
A sub-agent demonstration implementation with IBM Bob Introduction Large Language Model...

Claude Code subagents model selection made practical: when to give each subagent Haiku, Sonnet, Opus or inherit, with official prices, steps and fixes.
Anthropic's October 2026 guidance: fold many subagents into one loop with skills and MCP tools, migrate one at a time, and keep a subagent only when it earns its own context. A practical single agent vs multi-agent checklist.

A while back I wrote the skills guide I wish I’d had. Skills stop your agent from forgetting what it...
Sub-agents in Claude Code can carry MCP servers the parent session never sees. Configure mcpServers, tools, and disallowedTools in .claude/a

Anthropic's own production multi-agent research system spent roughly 15× more tokens than chat and...

Originally published at norvik.tech Introduction Explore the mechanics of Claude Code...
Skills, subagents, and MCP servers — three ways to extend Claude Code. When to use each, common mistakes, and how they compose together.
Part 2 of the Claude Code series: remote sessions, Agent SDK, subagents, hooks, and MCP integrations that 99% of users never find — with working examples.
Reach for a subagent when the task feels big, and reach for a skill for everything else — that's...