{"version":"2.1.285","anchor":"remote-tool-serving-capture-at-pipeline-end-and-allow-after","canonical_anchor":"remote-tool-serving-capture-at-pipeline-end-and-allow-after","heading":"Remote tool calls handle approvals that arrive after a stop more carefully","tier":"internal","area":"Elsewhere","url":"https:\/\/changelogs.core-directive.com\/v\/2.1.285\/e\/remote-tool-serving-capture-at-pipeline-end-and-allow-after","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.285","markdown":"### Remote tool calls handle approvals that arrive after a stop more carefully\n\nWhen Claude runs tools remotely, an approval that arrives after a stop is now recorded, and the end of the work waits for pending questions\n\n**Unclear.** It is not clear whether this changes anything a user would notice.\n\n**What**\n\nWhen Claude Code runs tool calls remotely, it now handles stops and permission questions more carefully. A tool call is an action Claude takes, such as running a command. If an approval for a tool call arrives after the stop signal has already been sent, Claude Code now records that it happened. The final step that collects the results is also delayed so it lines up with any permission questions still waiting for an answer.\n\n**Why**\n\nThis makes remote tool calls, such as those made in cloud sessions, less likely to go wrong when a stop and an approval cross paths.\n\n- Area: Elsewhere\n- Tier: Under the hood\n- Useful: 1\/5\n- Signal: 2\/5"}