Repository navigation
fix(rsr): criterion 3.2.2 is the verdict of the capability reader - #1188
Conversation
Owner decision 2026-11-16 (the elegant option of the two-reader finding from #1033): 3.2.2 now passes exactly when declared_capabilities/1 finds a profile, the reader behind the scorecard's profile_present. Before, with no deed (rsr-profile ...) clause, 3.2.2 passed whenever <machine tree>/rsr-profile.a2ml existed, even if it did not parse or carried no capabilities list, while the scorecard reported no profile. The RsrProfile model proved that this was the only disagreement (disagreementIsExactly, residualDivergence). The model follows the code: criterion322 = verdictOf . declaredCapabilities. Agreement becomes a theorem for every source (verdictAgrees, profilePresentAgrees); deedDecidedAgrees is now an instance of it; the fail-closed and fallback criterion theorems are re-proved over the reader lemmas; the old divergence becomes emptyA2mlFails, with tests/reject/EmptyA2mlPassesProfileCriterion.idr as its regression guard. ExUnit: a new case runs 3.2.2 and profile_present over an a2ml with a list, without a list, with the wrong section and unparseable. PROOF-STATUS rows for the changed model are OPEN until this commit's Verify Proofs idris2 --check job runs; the earlier PROVEN receipt covered the previous model. Not run locally: no Idris2 here, and hex.pm is blocked, so mix could not fetch deps. check-proof-status.sh, proof-ledger.sh check/claims and check-trusted-base.sh pass. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01E294yTkKk127qgKEXFFDBE
|
Warning Review limit reachedYou've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository. Next included review available in 30 minutes. View limit details
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. |
Codex Review Summary: ❌ FailedThis 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. |
|
This security review couldn’t start because the paying account has exhausted its Codex allowance and has insufficient available credits. The payer can check their balance and add credits in the Codex usage dashboard, then try again. If you do not manage the paying account, contact this repository's admins. |
…ESTED Verify Proofs run 43939522672, idris2 --check job 133552988403, at 99a40d6 (Idris 2 v0.7.0, ABI package build) checked the changed RsrProfile model: PASS: no escape-hatch marker in src/Hypatia/ABI/RsrProfile.idr PASS: accepted positive control RsrProfileResolution PASS: rejected EmptyA2mlPassesProfileCriterion at invalid: Mismatch between: Pass and Fail. PASS: ABI package, 11 positive control(s), 22 expected rejection(s) The model row is PROVEN on that receipt; the accept and the new reject fixture are TESTED. The Elixir correspondence stays ASSUMED. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01E294yTkKk127qgKEXFFDBE
|
This security review couldn’t start because the paying account has exhausted its Codex allowance and has insufficient available credits. The payer can check their balance and add credits in the Codex usage dashboard, then try again. If you do not manage the paying account, contact this repository's admins. |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: c9e8ce4400
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".
Codex review on #1188: 3.2.2 called declared_capabilities/1 a second time instead of using the `caps` score/2 had already read. A profile rewritten mid-scan (the delegated scan runs between the two reads) could then leave profile_present and the gated applicability on the old reading while 3.2.2 judged the new one, breaking the agreement the RsrProfile model proves, which assumes a single read. evaluate/4 now passes `caps` to a detector of arity 2, and 3.2.2 is such a detector: it judges exactly the value the scorecard reports. The other detectors are unchanged (arity 1). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01E294yTkKk127qgKEXFFDBE
|
This security review couldn’t start because the paying account has exhausted its Codex allowance and has insufficient available credits. The payer can check their balance and add credits in the Codex usage dashboard, then try again. If you do not manage the paying account, contact this repository's admins. |
|
This security review couldn’t start because the paying account has exhausted its Codex allowance and has insufficient available credits. The payer can check their balance and add credits in the Codex usage dashboard, then try again. If you do not manage the paying account, contact this repository's admins. |
…n-vsrzhw-322 # Conflicts: # CHANGELOG.adoc
|
This security review couldn’t start because the paying account has exhausted its Codex allowance and has insufficient available credits. The payer can check their balance and add credits in the Codex usage dashboard, then try again. If you do not manage the paying account, contact this repository's admins. |
|
## Change `HYPATIA_SHA` in `.github/workflows/hypatia-scan.yml` moves from `7c10e5d` to `6d8ab66`, hypatia's `main` head on 2026-11-16. The new scanner includes: - secret-scanning findings that name the secret's file and line, and are never hidden by path exclusions (hyperpolymath/hypatia#1184); - the RSR 3.2.2 verdict from the capability reader (hyperpolymath/hypatia#1188); - the parsed workflow model for WH013 (hyperpolymath/hypatia#1192); - the actions-lock proof ledger (hyperpolymath/hypatia#1187). The workflow's interface to the scanner is unchanged at that commit: `hypatia-cli.sh` is still there, and the build is still `mix escript.build`. ## Findings at the new pin From [run 43967735672](https://github.com/hyperpolymath/echo-types/actions/runs/43967735672) on `29278de`: - **Build.** The cache missed, so the scanner was built fresh from `6d8ab66` ("Generated escript hypatia"). - **Hypatia.** Both the JSON and SARIF scans report `0 findings >= medium (critical=0, high=0, medium=0, warn=0, low=0, info=0)`, and the counts step reports critical 0, high 0, total 0. The gating row is 🟢 clean. - **Alert-backed rules.** These read 0 observations. The job token can read code-scanning and Dependabot alerts but not secret-scanning alerts, which need an App or PAT. That limit is unchanged by this PR. The 🔴 rows in the rendered grid come from **advisory** fleet reporters at the pinned `gitbot-fleet` revision, not from Hypatia, and this PR does not change them: - **rhodibot (RSR):** 11 high. Most expect `.a2ml` descriptiles (`ANCHOR.a2ml`, `STATE.a2ml`, `0-AI-MANIFEST.a2ml`, `.machine_readable/descriptiles/` and others). A2ML is retired estate-wide, so these rules are stale. They are not answered by creating `.a2ml` files here; rhodibot's criteria need updating to deeds in gitbot-fleet. - **finishingbot:** 2 high and 174 medium, mostly docs that name files which no longer exist (`All.agda`, `Smoke.agda`, `EchoVsQuotient.agda`, …). This is a docs clean-up for its own PR. - **glambot:** 1 medium and 136 minor (line length, heading levels). 🤖 Generated with [Claude Code](https://claude.com/claude-code) https://claude.ai/code/session_01P4yFkxW5Heb8p2YW7ZYs9F Co-authored-by: Claude <noreply@anthropic.com>



Why
The Idris2 model added in #1033 found one case where RSR criterion 3.2.2 (profile present) and the scorecard's
profile_presentdisagree:(rsr-profile …)clause, andrsr-profile.a2mlexists but does not parse or carries nocapabilitieslist.In that case 3.2.2 passed on the file's presence, while the scorecard reported no profile. The model proved this was the only disagreement (
disagreementIsExactly,residualDivergence). The owner chose the elegant fix on 2026-11-16: 3.2.2 uses the reader.What
lib/rules/rsr_conformance.ex:profile_present/0is nowdeclared_capabilities(repo) == :none → :fail, so 3.2.2 andprofile_presentread the same capability set and agree by construction.src/Hypatia/ABI/RsrProfile.idr: the model follows the code, withcriterion322 = verdictOf ∘ declaredCapabilities.verdictAgreesandprofilePresentAgreesprove agreement for every source.emptyA2mlFailscovers the formerly divergent case, which now fails.duplicateFailsCriterion,listlessFailsCriterionandfallbackCriterionNoClause.deedDecidedAgrees, the bffd76a fix, is now an instance ofprofilePresentAgrees.check322,a2mlFileVerdict,a2mlDecidedAgrees,disagreementIsExactlyandresidualDivergence, because they described behaviour that no longer exists.tests/reject/EmptyA2mlPassesProfileCriterion.idr(new): a regression guard. It claims the old behaviour and must be refused atinvalid.tests/accept/RsrProfileResolution.idr: the residual cases are replaced byemptyA2mlFailsHere,emptyA2mlNotPresent,emptyA2mlAgreesanda2mlListPasses.test/rules/rsr_conformance_test.exs: a new ExUnit case checks 3.2.2 andprofile_presentover four a2ml files: one with a list, one without, one with the wrong section, and one that does not parse.PROOF-STATUS.adoc, the ABI table in.claude/CLAUDE.md, andCHANGELOG.adoc.Effect on scores: a repo with no deed clause and an
rsr-profile.a2mlthat declares no capabilities now fails 3.2.2. The scorecard already treated it as having no profile, so its gated criteria were already:na.Evidence
CI on
99a40d6: 91 checks passed, 8 were skipped and 1 was neutral. None failed.Verify Proofs (run 43939522672,
idris2 --checkjob 133552988403, Idris 2 v0.7.0) logged:Elixir:
3555 tests, 0 failures, 3 skipped (246 excluded). The log does not name test files, so I am inferring that the new case ran rather than reading it there.Receipt commit:
c9e8ce4records that run inPROOF-STATUS.adoc.Merge: the branch has
mainmerged in (39a1b9f), so the new head reruns CI.Not run locally:
mix test: hex.pm is blocked by the session's network policy.Out of scope, and not new to this PR: the Verify Proofs log warns
proof-ledger: src/Hypatia/ABI/RsrProfile.idr checks in lane idris2 and no ledger row names it. That warning dates from #1033: the machine ledger has no row forRsrProfile.idryet.🤖 Generated with Claude Code
https://claude.ai/code/session_01E294yTkKk127qgKEXFFDBE