Repository navigation
Conversation
The probe was written on 2026-08-01 on a branch that had already carried PR #727, and never reached a PR of its own. `evals/generic_parity.py`, `evals/run_generic_parity.py`, `tests/test_generic_parity.py` and the two case/witness fixtures existed only in one local commit while #715 stayed open. Rebased onto main. The original commit's `decision-framing.md` and `research-priors.md` halves are dropped: main's versions are the later rewrite #727 landed, and they already carry the strategy-class map the older text proposed. `tests/run_all.py` registers the new suite in main's three-tuple lane format. 61/61 offline suites pass. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Why this is a draft
#715 is marked "Research / Discussion only — product decision recorded, not an active implementation front", and says not to merge ahead of the reliability fronts in #27. This PR adds no product behaviour — it is opt-in evaluation tooling — but the merge call is the owner's, so it opens as a draft rather than presuming the gate.
What this recovers
The probe was written on 2026-08-01 on
codex/issue-715-generic-parity, the branch that had already carried PR #727, and never reached a PR of its own. It has been sitting only in a local commit ever since:evals/generic_parity.pyevals/run_generic_parity.pytests/test_generic_parity.pyevals/judge_episodes.py(two shared text-call helpers)evals/generic-parity-cases.json,evals/generic-parity-witnesses.jsondocs/eval-design.md,evals/EVALS.md,tests/run_all.pyrun_generic_parity.pyis an opt-in, host-side A01/A07/A10 probe for the no-book route. It is not an engine route, not a product-surface capture, and not a Generic Parity acceptance gate: its receipt always saysowner_unreviewed/not_granted, andcandidate_passis iteration evidence, never merge authority.--planmakes zero model calls.What was dropped in the rebase
The original commit also touched
skills/fomo-kernel/references/decision-framing.mdandresearch-priors.md. Both are dropped — main's versions are the later rewrite that #727 landed, and they already carry the strategy-class map the older text proposed (main renamed the section to## Research-aware strategy framingand added the #716 cross-route note). Keeping the commit's halves would have reverted that work.tests/run_all.pyregisters the new suite in main's three-tuple lane format (qa-eval), not the two-tuple format the original commit predated.Verification
python3 tests/run_all.py→ PASS: all 61 suites passed (group: all), including the newly registeredtests/test_generic_parity.py(6/6).Refs #715. Does not close it — the design question the issue tracks is unchanged by this PR.
🤖 Generated with Claude Code