16 lines
944 B
Markdown
16 lines
944 B
Markdown
|
---
|
||
|
|
name: Autonomous execution preference
|
||
|
|
description: User prefers Claude to make reasonable calls and continue rather than pause for clarifying questions, especially during long-running sessions
|
||
|
|
type: feedback
|
||
|
|
originSessionId: 2352cf59-de5a-4440-8480-69099e3a4cbb
|
||
|
|
---
|
||
|
|
When given a multi-step build task, do not stop for clarifying questions — make the reasonable call and continue. The user will redirect if needed.
|
||
|
|
|
||
|
|
**Why:** User explicitly stated this mid-session: "I am going to bed so you will have to act alone." They value forward progress over confirmation, even on architectural choices.
|
||
|
|
|
||
|
|
**How to apply:**
|
||
|
|
- Pick sensible defaults and call them out in the final summary
|
||
|
|
- If a decision is genuinely irreversible (deleting work, pushing to prod), still pause — autonomy ≠ recklessness
|
||
|
|
- Document assumptions in code comments or README so the user can audit on wake-up
|
||
|
|
- Don't ask "do you want me to continue?" — just continue
|