{"version":"2.1.290","anchor":"skill-and-bash-in-prompt-handling-tweaks","canonical_anchor":"skill-and-bash-in-prompt-handling-tweaks","heading":"Small changes to skills and to shell commands written into prompts","tier":"notice","area":"Skills","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/skill-and-bash-in-prompt-handling-tweaks","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290","markdown":"### Small changes to skills and to shell commands written into prompts\n\nShell commands in prompts are read differently, skill use is recorded by the skill's name, and a pull request skill no longer takes a proactivity level\n\n**Unclear.** None of these was followed through, so their effect for users is not known.\n\n**What**\n\nSkills are reusable sets of instructions Claude Code can load. Three small changes:\n\n- Shell commands written into a prompt are now read with a new parser instead of only having spaces trimmed from their ends.\n\n- When Claude Code records that a skill was used, it now records it under the skill's own name.\n\n- The prompt for a pull request skill is no longer built with a proactivity level.\n\n**Why**\n\nThese are small changes to how skills and pull request skill prompts behave.\n\n- Area: Skills\n- Tier: You'll notice\n- Useful: 1\/5\n- Signal: 1\/5\n- Scope: individual\n- Heads-up: no"}