[hyperpolymath]

Summary

Two changes so the catalogue reflects metadatastician/hvn-dehallucinator as it is now:

• The census chose a repository's status document by alphabetical order. A template-era PROOF-NEEDS.adoc therefore won over the real ledger beside it, and hvn-dehallucinator was catalogued against the wrong file.
• That repository's row was from the 2026-10-26 census. Since then it has gained Haruspex*.agda, two reject fixtures and the differential-test TLA+ files.

Changes

• scripts/census/build-rows.sh ranks status documents: PROOF-STATUS, then PROOF-NEEDS, then proof-debt, then the path. Committed rows change only when they are rebuilt, so other repositories that carry both documents keep their old pick until the next full census.

• I rebuilt one row, metadatastician/hvn-dehallucinator at f2f31f7, with census-repo.sh, summarise.sh and build-rows.sh, and merged it into catalogue-rows.json by repo. Order and the one-line canonical form are unchanged.

  	own files	fixtures	provers	declarations	status document
  before (2026-10-26)	9	3	agda 5, idris2 1, tla 3	108	docs/status/PROOF-NEEDS.adoc
  after (f2f31f7)	16	7	agda 10, idris2 1, tla 5	203	docs/status/PROOF-STATUS.adoc

• docs/catalogue/catalogue.adoc and docs/atlas/generated/* are regenerated from the rows.

• CHANGELOG.adoc gets an entry for both changes.

I did not run run-census.sh with a one-repo list: it would rewrite catalogue-rows.json and unreachable.json with that one repository only.

RSR Quality Checklist

Required

• Tests pass:
  • render-catalogue.sh | diff catalogue.adoc: no difference (the catalogue-check recipe, run by hand because just is not installed here);
  • bun scripts/atlas/census-atlas.mjs check: PASS;
  • bun test tests/atlas/census-atlas.test.mjs: 10 pass, 0 fail.
• Code is formatted
• Linter is clean
• No banned language patterns (bash, jq)
• No unsafe blocks
• No banned functions
• SPDX license headers present on all new/modified source files
• No secrets, credentials, or .env files included

As Applicable

• CHANGELOG updated

Testing

As listed above. Every row except hvn-dehallucinator's is still the 2026-10-26 census. A row count is not a proof: theorem truth and receipts stay in the home repository (docs/standards/AUTHORITY.adoc).

---

Generated by Claude Code

[coderabbitai]

Caution

Review failed

The pull request is closed.

• Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

Share

• X
• Mastodon
• Reddit
• LinkedIn

[hyperpolymath]

Two checks fail on this PR. Neither is caused by it, and both fail the same way on main:

• Dogfood Gate / Canon lockstep. It compares canon.lock with mathesis_chora.deed and reports "the template does not implement the canon at HEAD". This PR changes neither file. The latest three Dogfood Gate runs on main all fail: a3dbe73 (run 42592014015), 4a90f39 and 1058bb8. Fixing it means moving this repository's canon pin, which is the owner's call.
• estate-rules (tests/shape/repo_map_determinism_test.sh). It reports that the committed docs/architecture/REPOSITORY-MAP.adoc is stale and should be regenerated with just repo-map. The same test fails on a checkout of main at a3dbe73. The fix is to regenerate the map on main with bash scripts/gen-repo-map.sh .. I have not done that here: it would fold unrelated changes into this PR.

---

Generated by Claude Code

[chatgpt-codex-connector]

Codex Review Summary: ❌ Failed

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	❌ Failed 2026-11-09T15:52:47.366824Z We couldn’t complete this review. Comment @codex review to try again.	654de35	Draft marked ready

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" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud