How the agent loop works changedagent-sdk/agent-loop
Nearest release: v2.1.233, published 10 hours before this site recorded the change. Shown because the two are within 24 hours of each other. Nothing here says the release caused the edit.
Recorded here
Lines+2added
Lines−0removed
From line
145
where the diff opens
First seen
14 Aug 2026
this site's first read of the page
Recorded edits20to this page, all time
The whole hunk
from line 145, old and new numbered
/
from line 145
145145| **Discovery** | `ToolSearch` | Dynamically find and load tools on-demand instead of preloading all of them |
146146| **Orchestration** | `Agent`, `Skill`, `AskUserQuestion`, `TaskCreate`, `TaskUpdate` | Spawn subagents, invoke skills, ask the user, track tasks |
147147
148On the [models that don't get the task-tracking tools](/docs/en/agent-sdk/todo-tracking#model-availability), Claude Code provides `TaskCreate` and `TaskUpdate` only when you opt in.
149
148150Beyond built-in tools, you can:
149151
150152* **Connect external services** with [MCP servers](/docs/en/agent-sdk/mcp) (databases, browsers, APIs)
No line in this hunk matches that.