Fix terminal input prompts to use askQuestions in autopilot mode#309583
Merged
meganrogge merged 2 commits intomainfrom Apr 13, 2026
Merged
Fix terminal input prompts to use askQuestions in autopilot mode#309583meganrogge merged 2 commits intomainfrom
askQuestions in autopilot mode#309583meganrogge merged 2 commits intomainfrom
Conversation
Contributor
Screenshot ChangesBase: Changed (1)
|
| Before | After |
|---|---|
 |
 |
Contributor
There was a problem hiding this comment.
Pull request overview
This PR updates the runInTerminal tool’s runtime steering text so that when a terminal command appears to be waiting for interactive input, the model is always instructed to go through vscode_askQuestions (instead of silently guessing and using send_to_terminal directly in auto-approve/autopilot-like modes). This aligns terminal-prompt handling with the askQuestions tool’s autopilot behavior (auto-response + rendered completed summary).
Changes:
- Remove the auto-approve/autopilot-specific “do not ask the user; guess and send input” instruction path.
- Always emit guidance that requires
vscode_askQuestions, followed by one-at-a-timesend_to_terminal+get_terminal_outputbetween prompts (idle, input-needed, timeout, and background steering paths).
Show a summary per file
| File | Description |
|---|---|
| src/vs/workbench/contrib/terminalContrib/chatAgentTools/browser/tools/runInTerminalTool.ts | Unifies terminal input-needed guidance to always route through vscode_askQuestions, removing the prior silent-guess path for auto-approve/autopilot-like sessions. |
Copilot's findings
- Files reviewed: 1/1 changed files
- Comments generated: 0
mjbvz
previously approved these changes
Apr 13, 2026
# Conflicts: # src/vs/workbench/contrib/terminalContrib/chatAgentTools/browser/tools/runInTerminalTool.ts
benvillalobos
approved these changes
Apr 13, 2026
meganrogge
added a commit
that referenced
this pull request
Apr 17, 2026
…esults Fix eval regression by: 1. Restoring isAutoApproved branch at all 4 steering text injection sites so autopilot/benchmark mode uses direct send_to_terminal instead of forcing through askQuestions 2. Shortening all steering text from ~100 words to ~25 words per site 3. Shortening timeout wrapper text Addresses bb77ece (#309050) and 8a73ca2 (#309583) regressions.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Fix #309524
When a terminal command
waits for input in autopilot mode, the agent was instructed to silently guess answers and send them directly; the user never saw what was answered or had a chance to review. Now the agent always goes through
askQuestions, which in autopilot mode auto-responds instantly and renders a completed carousel summary in chat so the user can see exactly what values were provided on their behalf.