{"version":"2.1.309","anchor":"tool-name-telemetry-classifier-now-falls-back-safely","canonical_anchor":"tool-name-telemetry-classifier-now-falls-back-safely","heading":"Tools with unreadable names no longer break usage statistics","tier":"internal","area":"Telemetry","scope":"org","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.309\/e\/tool-name-telemetry-classifier-now-falls-back-safely","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.309","markdown":"### Tools with unreadable names no longer break usage statistics\n\nClaude Code now labels a tool whose name cannot be read as unreadable in its usage statistics, instead of failing\n\n**What**\n\nWhen it records usage statistics, Claude Code sorts each tool into a category, such as a skill, an MCP tool or a tool added while the session runs. If a tool's name cannot be read, it is now labelled unreadable instead of causing an error. The category labels now come from a single shared list, so their exact wording may have changed.\n\n**Why**\n\nA tool with an odd name can no longer cause an error in the code that records usage.\n\n- Area: Telemetry\n- Tier: Under the hood\n- Useful: 1\/5\n- Signal: 1\/5\n- Scope: org\n- Heads-up: no"}