Publish Astra and Fable planning study results - #71
Conversation
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 5191de1afb
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".
| <div class="header-inner"> | ||
| <a class="brand" href="#top" aria-label="Co-Evolution home"><svg viewBox="0 0 36 36" aria-hidden="true"><path d="M15 7 4 18l11 11M21 7l11 11-11 11"/></svg><span>co-evolution<span class="brand-dot">.</span></span><span class="brand-label">EVALS</span></a> | ||
| <nav aria-label="Main navigation"><a href="#results" class="active">Results</a><a href="#compare">Compare</a><a href="#methodology">Methodology</a><a href="#archive">Archive</a></nav> | ||
| <nav aria-label="Main navigation"><a href="#results" class="active">Results</a><a href="planning.html">Planning study</a><a href="#compare">Compare</a><a href="#methodology">Methodology</a><a href="#archive">Archive</a></nav> |
There was a problem hiding this comment.
On narrow mobile viewports, adding this fifth navigation item makes the coding page's header overflow horizontally. The max-width:760px rule in observatory.css gives the navigation only the viewport width minus 40px while retaining a fixed 27px gap; at 320px that leaves 172px for all five labels after gaps, which is insufficient even at the configured 11px font size. Allow the links to wrap, reduce the mobile gap, or collapse the navigation so the new Planning link does not push content off-screen.
Useful? React with 👍 / 👎.
Publish the recovered custom planning screen in the existing evaluation observatory's visual style. A dedicated planning page exposes all 33 workflows, independent Astra/Fable rubric scores, paired comparisons, per-brief coverage, costs and downloadable cited evidence. The main site links to it.
The screen is terminal partial: 194/198 plans and 381/396 judgments, 730/790 charged calls. The primary paired effect is +9.00 for Astra and -1.00 for Fable across five briefs. Missing results, recovery amendments, cost gaps and exploratory limitations remain explicit. No additional model calls were made, and coding data and archived editions are unchanged.
Validation: exported counts and all row means matched the source; six existing observatory tests passed; desktop/mobile screenshots, author filtering and detail navigation checked. Original evidence and ledger hashes matched during offline settlement.