ticket-99: evidence-first public portfolio presentation - #20
Conversation
Tests: pytest -q; focused docs/data-policy tests; run-log validation Artifacts: docs/assets/portfolio/validation_frontier.csv and .svg Docs: README, evidence note, changelog, progress, ticket, and run log
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: e92a47e209
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| <line class="gate" x1="852" y1="72" x2="852" y2="365"/> | ||
| <text class="small" x="812" y="68">0.50 gate</text> |
There was a problem hiding this comment.
In this SVG the bars use x=402 as zero and about 600 px per Sharpe unit, so the 0.5 tick is at x=702 and the SEC bar for 0.4736 ends near x=686; however the dashed 0.50 gate is drawn at x=852. In the README evidence graphic this visually labels a much higher threshold as the promotion gate and makes the strongest rejected candidate look far farther from 0.50 than the source CSV says, so the gate/label should be moved to the 0.5 tick or the axis should be rescaled.
Useful? React with 👍 / 👎.
| "started_at_utc": "2026-07-14T21:48:14Z", | ||
| "finished_at_utc": "2026-07-14T22:09:33Z", | ||
| "git_sha_before": "31fe55320b5bdf3fdea386de98e627ad15290d0c", | ||
| "git_sha_after": "31fe55320b5bdf3fdea386de98e627ad15290d0c", |
There was a problem hiding this comment.
This new run log records git_sha_after as the same base SHA as git_sha_before, but the reviewed commit containing the README/evidence artifacts is 5fb9d0c931f4d5f4083b5bf5894def7665c57795. Any later provenance check that follows the recorded after SHA lands on a tree that does not contain this ticket's published files, breaking the reproducibility contract for the new public evidence.
Useful? React with 👍 / 👎.
|
CI investigation: all three matrix jobs stop at the repository-wide |
1 participant
Summary
Validation
pytest -q— 129 passedNo raw or licensed rows, sealed holdout outcomes, secrets, or strategy identifiers are included.