Orchestrate subagents at scale with dynamic workflows changedworkflows
Nearest release: v2.1.251, published 10 hours before this site recorded the change. Shown because the two are within 24 hours of each other. Nothing here says the release caused the edit.
Recorded here
Lines+8added
Lines−0removed
From line
319
where the diff opens
First seen
14 Aug 2026
this site's first read of the page
Recorded edits46to this page, all time
The whole hunk
from line 319, old and new numbered
/
from line 319
319319
320320Every run writes its script to a file under your session's directory in `~/.claude/projects/`. Claude receives the path when the run starts, so you can ask for it. You can open that file to read the orchestration Claude wrote, diff it against a previous run's script, or edit it and ask Claude to relaunch from the edited version.
321321
322Claude can start a workflow from a script file only when it can already read that file without a permission prompt, such as:
323
324* The run's own persisted script under `~/.claude/projects/`
325* A file in your working directory or an [additional directory](/docs/en/permissions#working-directories)
326* A path your [Read allow rules](/docs/en/permissions#read-and-edit) cover
327
328Claude Code refuses a script path Claude can't read that way.
329
322330The runtime tracks each agent's result as the run progresses, which is what makes a run [resumable](#resume-after-a-pause) within the same session.
323331
324332### Prompt caching in a fan-out
No line in this hunk matches that.