Repository navigation
Keep the prompt out of Claude Trials' disallowed tools - #1337
Merged
Merged
Conversation
--disallowedTools takes every following word as a tool name, so with its value as a separate argument it also took the prompt after it, and Claude exited: "Input must be provided either through stdin or as a prompt argument". Every Claude Trial, model Grader, Scoring Grader and Bug-origin session ran with no prompt. Pass the flag and its value as one argument. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01LSAzXiA6BD9neiPgVQRMKD
JacobStephens2
added
the
test:macos
1 task
Merged
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Every private session on Claude Code exited at once:
That covers Claude Trials, model Graders, Scoring Graders and Bug-origin proposals. Found on the
thirdshifthost today: #1255's live check had its model Grader fail in 8 seconds, and the first Scoring run (#1032) failed in its Bug-origin step.Cause:
Claude::trial_sessionput--disallowedToolsandWebSearch,WebFetchas two arguments just before the prompt. Claude Code declares that flag as<tools...>, so it took the prompt's words as more tools to deny:Fix: pass
--disallowedTools=WebSearch,WebFetchas one argument. The unit test now requires that form, and it failed before the change. The fake Harness doesn't parse variadic flags, which is why the integration tests missed this.No other Claude invocation puts a variadic flag before its prompt: the summary's
--tools ""is followed by a flag, and its prompt goes on stdin. Grok's--disallowed-toolstakes a single value.🤖 Generated with Claude Code
https://claude.ai/code/session_01LSAzXiA6BD9neiPgVQRMKD