TLDRocket
Sign in

Anthropic’s Claude Code gained “Claude Mods,” allowing JavaScript/TypeScript plugins to customize the agent’s harness behavior

Feature update Provisional 66% confidence first seen

The coverage describes “Claude Mods,” a capability for Claude Code that lets users install JavaScript/TypeScript plugins to modify how the agent handles prompting, tool calls, and rendering. It also highlights built-in mods (including one that checks for uncommitted edits) and a “Next Steps Supervisor” behavior that activates after repeated tool calls, changing the agent’s attention/compute loop.

Claude Code Claude Mods Anthropic Attention Interface Claude Sonnet Next Steps Supervisor You Should Know

Source coverage

Claude Mods: Flipping the Harness Inside Out Latent Space newsletter Claude Mods: Flipping the Harness Inside Out Latent Space newsletter

The daily briefing

Every AI story that matters, in your inbox by 8am.

TLDRocket reads all relevant sources, removes duplicate coverage, and summarises the day in two minutes. Follow companies and topics for alerts, or get the briefing in Slack. Free, no spam, unsubscribe anytime.