34 posts tagged with "mechanics"

Claude Code cross-session messaging lets your sessions message each other. How ListAgents and SendMessage work, inbound controls, and the limits.

Anthropic cut 80% of Claude Code's system prompt. We ran the same surgery on a production framework: 8,798 agent lines to 664. Here is the diff.

Every CLAUDE.md guide tells you what to add. After Claude 5, subtraction wins. The delete list, the one test per line, and what must never be cut.

Claude Code runs four loop types: turn-based, goal-based, time-based, and proactive. What each means and how to set a real stop condition.

What Claude Code /compact keeps and loses, straight from Anthropic's docs. The full survival table plus how to make rules and skills persist.

Claude Code Artifacts turn session work into live, shareable pages at a versioned URL. Private by default, in beta for Team and Enterprise.

Skills can be scoped to file paths. Workflows load only where they apply. Here is how, plus the rules vs workflows mental model.

One CLAUDE.md does not scale. Layer them per subdirectory. Claude walks the tree, loads the right rules, and your context stays lean.

Claude Code is a model plus a harness. The AI layer of skills, hooks, agents, and MCPs matters more than the model. Here is how it works.

Claude Code voice mode lets you hold spacebar to talk, release to transcribe. Mix typed and spoken input in a single prompt.

Claude Code 2.1.129 added skillListingBudgetFraction, silently dropping skills past 1% of context. Exact settings.json format to raise it.

Claude Design hands prototypes off directly to Claude Code. The only AI design tool with a closed design-to-production loop. Walkthrough inside.