Source Intelligence
Follow Discord

DisclaimerUnofficial, and not affiliated with Anthropic. Nearly all of this is read straight out of what ships: npm bundles, captured prompts, published docs. Anthropic's own notes go in verbatim, marked as theirs. The rest is my reading, and every entry carries the strings behind it. If one looks wrong, vote it down and say why.

← All of v2.1.187 Home All releases ←olderv2.1.186 v2.1.190newer→
Claude Code v2.1.187

Skill token footprint in /usage

Plugin pages in /usage now show how much of the system prompt context budget each plugin's skills consume, broken down per skill.

Details
  • Shown under "Skill-listing footprint" — the cached input cost after the first turn, before agents and MCP tools are counted
  • Lists each skill with its approximate token cost per turn, sorted by size
  • Footer note: "For per-skill invocation counts and cost attribution, see /usage"
  • When no model-invocable skills are loaded for a plugin, shows "No model-invocable skills loaded for this plugin"
Evidence

New skill footprint UI (search for "Skill-listing footprint" or "What this plugin's skill descriptions add to the system prompt")

Strings lifted out of the shipped bundle, so the claim above can be checked against them.

See this entry in the whole of v2.1.187 →