Author(s): Rajesh Pandhare Originally published on Towards AI. The Month Claude Code Became a Fleet and Got BrakesWhat 22 July releases (v2.1.198 → v2.1.220 ...
Anthropic's background subagents launched without budget brakes (July 1); `--max-budget-usd` failed to halt fleets for 20 days. Unattended CI/overnight automation now requires explicit spend caps and tool restrictions to avoid invisible cost spike.
Author(s): Rajesh Pandhare
Originally published on Towards AI.
What 22 July releases (v2.1.198 → v2.1.220) tell you about where AI-assisted development is going, and the five limits you should tune today.
Buried in the July 21 release notes for Claude Code v2.1.217 is this one-line bug fix:
“Fixed --max-budget-usd not stopping background subagents: once the cap is reached, new spawns are denied and running background agents are halted."
Claude Code’s July 8 changelog is a useful reminder of what production agent engineering actually...
I let Claude Code build features, debug production issues, and refactor code for a month. Here's what it's actually good at.
Issue #63 | What 500,000 lines of production agent code teach us about building and optimizing agents
Author(s): Rick Hightower Originally published on Towards AI. Part 6 Subagents: Subagents are the Claude Code feature most…
Author(s): allglenn Originally published on Towards AI. Core concepts: what “usage limits” actually measure You’re forty minutes…
Claude Code 2.1.232 changes more than the version number. Forked subagents now inherit the full...
