Replace the /examples gallery with four audited public pull requests - #5
Merged
Conversation
…uests The old gallery was built on a run of the analysis whose output has since been corrected, and several of the checks its findings were attributed to no longer exist. Everything on the page is replaced with material from a recent scan, and every rendered claim was checked by hand against the two revisions it names before it was allowed on. The four, and what each shows that a file diff cannot: smallrye/smallrye-graphql#1424 a rename removes two methods from a type twelve components depend on citrusframework/citrus-simulator#370 one new edge closes a loop that runs config -> scenario.mapper -> http -> listener smallrye/smallrye-graphql#2083 a two-file change crosses out of the client module into the server API package spring-ai-community/spring-ai-session#33 nineteen sentences from the repo's own docs, checked against the code, all holding Each card now carries a "what we checked" column: the base and head SHAs, a link to the compare view, and the reproducible steps behind every claim. The fabricated GitHub chrome the old page wrapped these in - invented tab counts, a made-up check suite sidebar - is gone; it was a fake record of somebody else's pull request. Three otherwise strong candidates were rejected in the audit and are not here: one whose review note asserted callers would fail to compile when the PR had in fact updated them all, and two documented-rule rows whose subject had been bound to the wrong type. The diagram is part of the claim - a rendered SVG carries its review notes inside it - so an unverifiable note disqualifies the whole image, not just the text beside it. The page also states the sample it was drawn from: 79 public pull requests across 27 repositories, 45 of which produced no finding at all. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017X3m5Qxtn8KZPM6TqrknGk
✅ Deploy Preview for striff-io ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
`fitOnMobile` scales the whole diagram to the stage width and centres it, so a wide diagram ends up shorter than the stage and sits in a band of empty grid. The two landscape diagrams get mobile stage heights close to their fitted height; the two portrait ones keep theirs, since those are taller than any usable stage and are cropped rather than padded. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017X3m5Qxtn8KZPM6TqrknGk
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.
Replaces the
/examplesgallery. The old one was built on a run whose output has since been corrected, and several of the checks its findings were attributed to no longer exist.The four examples
config → scenario.mapper → http → listener → configFour kinds of finding, four repositories, four very different diagram scales.
The audit
Every rendered claim was checked against the repository at both revisions — for each example the base SHA was confirmed to be the pull request's merge base, so nothing shown is an artefact of a stale branch. Each card carries a "what we checked" column on the page: the two SHAs, a link to the compare view, and the reproducible steps.
Three otherwise strong candidates were rejected:
SessionService, the rule saysSessionRepository).A rendered SVG carries its review notes inside it, so an unverifiable note disqualifies the whole image, not just the text beside it. That is why the first rejection took the diagram with it.
Also
The fabricated GitHub chrome the old page wrapped the diagrams in — invented "Conversation 22 / Checks 11 / Files changed 17" tab counts, a made-up check-suite sidebar — is gone. It was a fake record of somebody else's pull request.
The page states its sample: 79 public pull requests across 27 repositories, 45 of which produced no finding at all. Four hand-picked examples say nothing about how often any of this fires, and the honest answer is: usually nothing.
Not in this PR
src/pages/index.astrostates the check count wrongly in several places (eighteen structural / nineteen total; the true roster is 14 structural + 2 documented-rule = 16, verified againststructuralCheckNames()anddocCheckNames()on the API'smaster). Those sentences exist only in the working tree's uncommitted edits to that file, so the corrections cannot be committed without also committing that in-progress work. The fix is applied in the working tree, uncommitted.🤖 Generated with Claude Code
https://claude.ai/code/session_017X3m5Qxtn8KZPM6TqrknGk