Repository navigation
docs: a README opening that says what this is, and the companion cross-links - #56
Merged
Merged
Conversation
…s-links
The lead paragraph was a feature list. The first thing somebody sees is now a
claim they can evaluate ("unit tests for the part of your app that has no right
answer"), a thirty-second copy-pasteable loop ending in real gate output, and a
five-row table of what this does that other eval tooling stops short of --
each row linking to the guide that explains why.
Additive only: nothing was removed, the table of contents gained entries rather
than losing any, and the README is 130 lines longer than before.
Also adds the companion-packages section the last two features earned:
- eval-harness-ai-bridge evaluates a laravel/ai agent -- the translation lives
there and not here, so an eval written today survives a move to another
runtime.
- laravel-evidence-risk-review ships an evidence-risk metric that is
deterministic and costs nothing per row, so it runs on every row of every
build next to the judge you can only afford on some of them.
Both are cross-linked from the guides they belong to, and the metric count is
now coherent across README, docs-site and composer.json: fifteen answer metrics
plus seven trajectory metrics is twenty-two.
Co-Authored-By: Claude Opus 5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01EZ4u82zdoa8naYm75kppMC
Deploying laravel-patent-box-tracker with
|
| Latest commit: |
0dee424
|
| Status: | ✅ Deploy successful! |
| Preview URL: | https://ec56bc5b.laravel-patent-box-tracker.pages.dev |
| Branch Preview URL: | https://claude-readme-and-crosslinks.laravel-patent-box-tracker.pages.dev |
Deploying eval-harness with
|
| Latest commit: |
0dee424
|
| Status: | ✅ Deploy successful! |
| Preview URL: | https://f64650f5.eval-harness.pages.dev |
| Branch Preview URL: | https://claude-readme-and-crosslinks.eval-harness.pages.dev |
![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: ece1a95592
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".
tests/Architecture/StandaloneAgnosticTest forbids sibling-package names anywhere under src/, prose included, so a real coupling cannot arrive dressed as a comment. A P3a docblock named padosoft/laravel-flow as an example of where an approval id might come from, which broke the invariant and turned main red -- on the Architecture suite, which `vendor/bin/phpunit` does not run by default (defaultTestSuite="Unit") while CI runs both. The example is now unnamed, with a note saying why so the next person does not helpfully put it back. Unit 1136 green, Architecture 3 green. Co-Authored-By: Claude Opus 5 <[email protected]> Claude-Session: https://claude.ai/code/session_01EZ4u82zdoa8naYm75kppMC
4 of 8 tasks
lopadova
added a commit
that referenced
this pull request
Aug 25, 2026
…r printed (#57) Three Codex threads on #56, all real. Copy-pasting the thirty-second version into a fresh install produced three different errors, none of them visible to someone reading the page. - The registration snippet had no home and no system-under-test. `eval-harness:run` starts its own Artisan process, so a bare `$eval->dataset(...)->register()` reaches "Dataset is not registered", and without a callable or SampleRunner under `eval-harness.sut` it reaches "No system-under-test bound". Shown as a registrar, with the binding, matching the docs-site quickstart. - `--compare=baseline --max-regressions=0` was the first command run. On an installation with no baseline, resolveReferenceReport() warns and skips the comparison — so the gate the surrounding paragraph is selling cannot fire, and the run passes. There are now two runs: the first records the baseline with --promote-baseline, the second is gated against it. - `brief report.json` read a file the preceding run never wrote: without --json --out the report goes to stdout as Markdown. The gated run now writes run-2.json and comparison.json on the reports disk, which is where BriefCommand resolves a relative path. The illustrated output was wrong in four smaller ways while I was in there: the `Macro-F1 …, pass rate …` headline belongs to RunBriefing, not to the run; the comparison line prints a trailing `%d compared`; the gate message is "N confident row(s) regressed against <reference> (allowed: N)"; and the counts claimed two improved rows in a one-sample dataset. Read off the format strings and moved the headline down to the briefing step where it is emitted. Added a note that llm-as-judge needs a provider and rouge-l does not, so the loop is runnable with no API key. Two LESSON entries. Unit 1136, Architecture 3, Pint and composer validate --strict clean. Claude-Session: https://claude.ai/code/session_01EZ4u82zdoa8naYm75kppMC Co-authored-by: Claude <[email protected]>
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.
The lead paragraph was a feature list, which is what a reader gets after they have decided to care. The first thing they see is now:
Additive only
Nothing was removed. The table of contents gained entries rather than losing any, every existing section is untouched, and the README is 130 lines longer than before.
Companion packages
The two features that shipped outside this repository now have the cross-links they earned, in a new section and in the guides they belong to:
eval-harness-ai-bridgelaravel/aiagent — the SDK translation lives there and not here, which is exactly what lets an eval written today survive a move to another runtimelaravel-evidence-risk-reviewevidence-riskmetric that is deterministic and costs nothing per row, so it runs on every row of every build next to the judge you can only afford on some of themBoth illustrate the same rule, stated in the section:
MetricResolverresolves any FQCN through the container, so a host package ships a metric withpadosoft/eval-harnessas arequire-devand nothing registers on either side.One correctness fix
The metric count was stale in three places. Fifteen answer metrics plus seven trajectory metrics is twenty-two, and the README, the docs-site description and
composer.jsonnow agree on that.Verification
composer validate --strict,phpunit(1136 tests), PHPStan level 6 and Pint all green. No source file changed.🤖 Generated with Claude Code
https://claude.ai/code/session_01EZ4u82zdoa8naYm75kppMC
Generated by Claude Code