Change to file-based output capture to handle large command outputs#33
Change to file-based output capture to handle large command outputs#33
Conversation
Co-authored-by: eriksw <24530198+eriksw@users.noreply.github.com>
|
@copilot The error caused by passing large outputs was incorrectly shortened and should have been quoted verbatim. The PR only describes the changes to stdout but both stdout and stderr are changed to output to files. |
You're correct on both points. The error message should be the full verbatim error, and the PR description should clarify that both stdout and stderr outputs are changed to file-based outputs (stdout_file and stderr_file), not just stdout. The README.md correctly documents both changes. The PR description will be updated to reflect these corrections. |
|
@copilot There are failing checks, fix the issues. |
Co-authored-by: eriksw <24530198+eriksw@users.noreply.github.com>
Plan: Capture outputs to files instead of GitHub outputs
This is a breaking change to address the "Argument list too long" error when passing large outputs to subsequent steps.
Changes Required:
executeCommand()to write stdout/stderr to files instead of capturing in memory.pipe()to stream output to both files and process.stdout/stderrstdout/stderrtostdout_file/stderr_filecontaining file pathsaction.ymlto reflect new output namesREADME.mdwith breaking change notice and new usage examplesOriginal prompt
💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.