feat(experiment): preflight_only credential-check mode — verify secrets with zero spend - #69
Merged
Conversation
The owner cannot visually confirm a GitHub Secret's value (write-only by design), and the mock run never touches credentials — so before spending on a real run there was no way to prove OPENWEIGHT_API_KEY + OPENWEIGHT_BASE_URL actually work. New workflow input preflight_only: one ~8-token probe call per judge, no scoring, no files, no spend. Exit 0 only when EVERY configured judge answered (green run = launch the real run; red run = the log names the exact judge/secret to fix). run-002 README gains the credential-check step and documents that the 'Updated X ago' timestamp + preflight are the only confirmations that exist for a secret. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Y7wiMwFa2D4P8zD9RhCgsv
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Prove the secrets work BEFORE spending on a run
The owner cannot visually confirm a GitHub Secret (values are write-only by design — the empty field on the edit page is normal), and the mock run never touches credentials. Run-002 attempt 1 showed what that blind spot costs.
New workflow input
preflight_only: one ~8-token probe call per judge, no scoring, no files, no spend. Exit 0 only when EVERY configured judge answered:✗ open-weight: preflight call failed: … 401 …).The run-002 README gains this as the credential-check step and documents the only two confirmations that exist for a secret: the "Updated X ago" timestamp and this preflight.
Validation: web 414/414, cold tsc clean. Dispatch-only workflow path.
🤖 Generated with Claude Code
https://claude.ai/code/session_01Y7wiMwFa2D4P8zD9RhCgsv
Generated by Claude Code