Repository navigation
Proof receipt from the shared generator, at 15fa113; gate it in CI (31 rows PWG-2ii) - #52
Conversation
The receipt now describes source commit 2b9bf3f and proofs.yml run 44202493421. C-4b gets the same three legs as C-4a. M-2 gets a Lean leg (LEAN-CHECKED), an unavailable Rocq leg and an open Agda leg, since the receipt profile gives every claim an Agda source leg. The Lean job checks the leg and the M-2 bridge in one run, so all Lean legs share one receipt, which imports tropical-types at f2e465c (PROOF-POLICY 8.5.1). The Rocq receipts name their digest-pinned image. The reader at standards 940a27af derives PWG-2ii for 28 rows and ungraded for M-2. No row is PWG-1: since standards#1625, PWG-1 needs every counted checker to run in an image pinned by digest, and the Agda and Lean legs do not yet. The earlier PWG-1 grades came from a reader that predates that rule; the proofs are unchanged. PROOF-NEEDS gains PN-6 for the images. check-proof-receipt.sh holds for every check, the import included. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01LiQzWJdcPfRU6YMDGRG8Xi
The Receipts table gave the Agda and Lean receipts' :inputs-sha256 but the file SHA-256 of ResourceAlgebra.v for Rocq. It now gives the Rocq receipt's :inputs-sha256 (608d32f3…c6eb296a5), with the file hash labelled. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01LiQzWJdcPfRU6YMDGRG8Xi
Main merged #45 (E-1, E-2, E-3: the Evidence algebra and ledger status derivation, with Rocq and Lean legs). Rather than land a receipt for 2b9bf3f that leaves three rows ungraded, the receipt now describes 15fa113, the merge of #45, from proofs.yml run 44209774341, in which all three legs and the M-2 bridge passed. - The Agda manifest adds Instances/Evidence.agda and Ledger.agda. - E-1, E-2 and E-3 get the same three legs as C-4a. - Every Lean receipt still lists the bridge with ResourceAlgebra.lean and lean-toolchain and imports tropical-types at f2e465c (one input digest per commit and prover). - The reader at standards 940a27af derives PWG-2ii for 31 rows and ungraded for M-2 (32 rows); check-proof-receipt.sh holds for every check, the import included. PROOF-STATUS conflicts: the census is restated for 15fa113. The E-1 row is kept from main and graded, and the recheck-legs note no longer excepts the E rows. PROOF-NEEDS: main's PN-6 (E-row recheck legs) stays and records the grade; the digest-pinned image item becomes PN-7. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01LiQzWJdcPfRU6YMDGRG8Xi
Conflicts in docs/status only. PROOF-STATUS census: 32 theory rows, 28 PWG-2ii at 2b9bf3f and 4 ungraded (M-2, and E-1..E-3, which postdate the receipt's commit). PROOF-NEEDS: main's PN-6 (recheck E-1..E-3, done in #45) keeps its number, and the digest-pinned-images need this branch added as PN-6 becomes PN-7; the CHANGELOG reference follows. Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_013CuBToxr5h9catptGiTsFX
scripts/proof-receipt.sh is this repository's home script for the estate's shared receipt generator (standards scripts/proof-receipt.sh): it holds the claim list, the run ledger and the hooks that name each leg's invocation and input set, and fetches the generator, the reader, the reference digest and the git checker at one pinned standards commit, each checked against its sha256. The receipt is now written by --write and checked by --check, not by hand. The receipt describes 15fa113 and proofs.yml run 44209774341, in which all three legs passed. The reader derives PWG-2ii for 31 rows, among them C-4b and E-1, E-2, E-3, which had no receipt, and ungraded for M-2 (a Lean bridge with no Agda statement). No row is PWG-1: only the Rocq job runs in an image pinned by digest (PN-7). The Lean input set gains scripts/check-lean-bridge.sh, which holds the pin of tropical-types and says of itself that a repin changes the home input digest. Before this, the generator was run against the hand-written receipt at 2b9bf3f: the reader's output was identical for all 29 claims. The standards pin (f4e83044) is on the branch claude/receipt-open-claims, not yet on standards main. Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_013CuBToxr5h9catptGiTsFX
Add a `receipt` job that runs `scripts/proof-receipt.sh --check` with a pinned Bun (PROOF-POLICY section 8.5.1 and 8.5.3), and add it to the `Proofs verdict` job's needs. Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_013CuBToxr5h9catptGiTsFX
PROOF-STATUS, PROOF-NEEDS and CHANGELOG now describe the receipt at 15fa113 (merge of #45), run 44209774341: 32 theory rows, 31 PWG-2ii, 1 ungraded (M-2). E-1 to E-3 and C-4b are PWG-2ii. The receipt is generated and checked by scripts/proof-receipt.sh; recording a new run is still manual. The receipt at 2b9bf3f joins the superseded list. Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_013CuBToxr5h9catptGiTsFX
… asks of a change The digests in the receipt table are the generator's, from git objects with the reference script, not the run summaries' commands. A change to a leg's inputs now fails the receipt job until the ledger marks that leg's row superseded and the receipt is regenerated; say so where a reader will meet it. Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_013CuBToxr5h9catptGiTsFX
…tor branch #50 merged main itself and rewrote its hand-written receipt for 15fa113 and run 44209774341, the same commit and run this branch's generated receipt describes. Both give 31 rows PWG-2ii and M-2 ungraded, and both number the digest-pinned-images need PN-7. The four conflicts (CHANGELOG, PROOF-NEEDS, PROOF-STATUS, the receipt) are resolved to this branch's side: the generated receipt and the documents that describe the generator and its gate. Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_013CuBToxr5h9catptGiTsFX
|
The suite's gate ( (run 44214954315). So every start of one of those workflows posts a The fix belongs in the suite, which I cannot push to. Proposed patch for the line that builds the description (tested with -desc=$(jq -rn --arg n "$RUN_NAME" '"re-run started: " + $n | if length > 140 then .[0:139] + "…" else . end')
+desc=$(jq -rn --arg n "$RUN_NAME" '"re-run started: " + ($n | [explode[] | select(. < 65536)] | implode | sub("^\\s+"; "")) | if length > 140 then .[0:139] + "…" else . end')I have not renamed the workflows here: that would be a downstream workaround for a suite bug, and their names come from the template. Generated by Claude Code |
|
State at 05:15 UTC: Still do not merge yet: the standards pin in
Generated by Claude Code |
|
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 44 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. |
|
…s not close PN-7) (#53) > **Re-scoped on 2026-11-18.** This PR first said it made the Agda and Lean jobs "isolated". The owner's ruling recorded in #51 (PN-7) says a digest-pinned base image with the checker installed at run time is **not** isolation: the legs earn `:image` only through prover images that ship the checker, from the planned `prover-images` repository. So this PR no longer claims that, and it does not close PN-7. What it does deliver is below. ## Summary - **Lean under lake.** PROOF-POLICY §8.2 names `lean` under `lake` as the qualifying Lean run (owner ruling 2026-11-17). The Lean leg and the M-2 bridge now build as Lake packages, with `lake-manifest.json` committed as a pinned input (§8.5.1). - **Base images pinned by digest** for the Agda and Lean jobs. This is hygiene: the runner image under the job no longer changes without notice. It is not the isolation §8.2 asks for, and the receipts name no image for these legs. - **The receipt follows the change.** Moving the Lean leg under lake changes that leg's inputs and its checker, so the receipt gate (from #52) refuses the old Lean run. The ledger marks it `superseded` and the receipt is regenerated. ## Grades Derived by the pinned reader: **31 rows PWG-2i, M-2 ungraded** (of 32). The Agda and Rocq receipts at `15fa113` stand; the Lean legs have no current receipt. After this merges and main's `proofs.yml` run passes, a follow-up records that run and the rows return to PWG-2ii. No row can be PWG-1 until PN-7 closes. ## Changes - `verification/proofs/lean4/lakefile.toml`, `lake-manifest.json`: a package with no dependencies, root `ResourceAlgebra`. - `scripts/check-lean-leg.sh` (new): builds a copy with `lake build`; fails on a failed build, on any `depends on axioms` report, on a count of axiom-free reports that differs from the `#print axioms` commands in the copy lake built, and on a manifest lake rewrote (it prints the difference). - `verification/bridges/tropical-types/lakefile.toml`, `lake-manifest.json`; `scripts/check-lean-bridge.sh` builds the three modules with `lake build`. The tropical-types pin and its fetch are unchanged; there is no lake `require` on that repository. Counts and the bridge digest come from the copy lake built. - `tests/check_lean_leg_test.sh` (new, 5 planted failures) and `tests/check_lean_bridge_test.sh` (6). - `.github/workflows/proofs.yml`: Agda in `debian:trixie-slim@sha256:a29215f6…`; Lean in `ubuntu:24.04@sha256:73b99939…` with the sha256-checked Lean 4.35.0 archive, `lean` and `lake` versions asserted; run summaries say the base image is pinned and the checker installed at run time; a digest that cannot be computed fails the step. - `scripts/proof-receipt.sh`: Lean row `superseded` with its reason; Lean input set gains both lakefiles, both manifests and `scripts/check-lean-leg.sh`; Lean invocation is the two lake builds. Receipt regenerated. - `docs/status/PROOF-STATUS.adoc`, `docs/status/PROOF-NEEDS.adoc` (PN-7), `CHANGELOG.adoc`: the census, the superseded Lean run, and the corrected wording on isolation. - `.gitignore`: `.lake/`. ## Testing Local, with Lean 4.35.0 and Lake 5.0.0: - `scripts/check-lean-leg.sh`: `PASS: Lean leg under lake; 84 of 84 printed theorems depend on no axiom`. - `scripts/check-lean-bridge.sh`: `PASS: M-2 bridge to hyperpolymath/tropical-types@f2e465c…; 6 of 6 printed theorems depend on no axiom`. - The two tests: 5 of 5 and 6 of 6 planted failures refused for the stated reason. - `scripts/proof-receipt.sh --check`: exit 0, ends `PASS: negative control (the first claim recorded as :pwg PWG-1 is refused: its legs support PWG-2i)`. - `scripts/check-proofs.sh lean4`, `tests/workflows/ci_gate_test.sh`, `check-ijson-jcs.sh --enforce` (14 of 14 canonical), `check-adoc-renders.sh` (164 files), `check-no-placeholders.sh`: pass. - `actionlint`: only the two SC2016 notices that are on main. In CI, on an earlier head (`d3e4651`), the Agda job, the Lean leg and bridge under lake, Rocq and `Proofs verdict` all passed in their containers, and `CI/gate` reported all 16 required checks passed. The receipt job did not exist on that head; it arrived with #52. ## Review CodeRabbit's finding that the Lean row must be superseded is done here. Its six smaller findings are fixed in `d7213f8` and their threads resolved. ## Not this PR's - The Hypatia scanner's intermittent segfault (see the comment below): it is not among the required checks. - CI failures when the repository's Actions API quota runs out; one re-run clears them. 🤖 Generated with [Claude Code](https://claude.com/claude-code) https://claude.ai/code/session_013CuBToxr5h9catptGiTsFX <!-- This is an auto-generated comment: release notes by coderabbit.ai --> <!-- coderabbit-ai-generated:v1 --> ## Summary by CodeRabbit - **Verification** - Updated Lean proof and bridge checks to build with Lake and verify axiom-free theorem reports. - Added checks that detect build failures and unexpected changes to dependency manifests. - Updated verification workflows to record image, tool-version, and input-hash details. Runtime installation of Agda and Lean checkers still does not meet the stated image-isolation requirement. - **Documentation** - Updated proof-status records to mark the previous Lean receipt as superseded; a qualifying Lake-based run is still pending. - Clarified current proof grades and the remaining image-isolation requirement. <!-- end of auto-generated comment: release notes by coderabbit.ai --> --------- Signed-off-by: Jonathan D.A. Jewell <6759885+hyperpolymath@users.noreply.github.com> Co-authored-by: Claude <noreply@anthropic.com>
…-unreconciled legs; reader: foundation-diverse needs RECONCILED (#1671) ## Summary 1. `scripts/proof-receipt.sh`: new optional hook `receipt_run_scope` (rows `prover|run|claim ids`), so a superseded run that checked only some claims is listed under those claims only. Needed by hyperpolymath/januskey#186. 2. A merge of main into the branch, and a test fix: the scope test's run rows now pass `--axioms none` and `--image`, which #1656 and #1625 made necessary. 3. Found by generating hyperpolymath/resource-algebra's receipt with the shared generator and comparing the reader's output with a hand-written receipt (28 of 29 claims identical; M-2, a Lean bridge with no Agda statement, differed): - Generator: a claim whose token is not `proven` or `proven-with-postulates` is now written `AGDA-PENDING` with no receipt (it was `AGDA-VERIFIED` carrying the Agda runs). - Generator: a recheck leg in state `COQ-CHECKED`, `LEAN-CHECKED` or `BRIDGE-PENDING` now gets token `proven` (`proven-with-postulates` when its run reported an axiom), not `open`, per PROOF-POLICY §2.2. It still counts toward no grade. `COQ-CHECKED` is refused on a non-Rocq leg, `LEAN-CHECKED` on a non-Lean leg. - Generator: optional `RECEIPT_BEHOLDING_CHORA` (`#u8"…"` or `#uuid"…"`), written after `:canonical-name`. - Reader (`deed_lint.js`, `readProofReceipt`): §8.3 gives the foundation-diverse tag to a kernel-checked witness whose correspondence is recorded as in §2.2. The reader now also requires state `RECONCILED`. - With these, the generator reproduces all 29 claims of resource-algebra's receipt reader-identically. 4. `1-formats/deed/tools/README.adoc`: one sentence stating the RECONCILED condition for the tag. PROOF-POLICY §8.3/§8.5 already say this and needed no change. Waiting on this: hyperpolymath/januskey#186 and hyperpolymath/resource-algebra#52. Note the branch head is now `340f546` (docs commit on top of `f4e8304`), so pins taken at `f4e8304` still hold the code unchanged. ## Tests (real output) - `bash scripts/tests/proof-receipt-test.sh`: `131 passed, 0 failed` - `bash scripts/tests/check-proof-receipt-test.sh`: `20 passed, 0 failed` - `cd 1-formats/deed/tools && bun test deed_lint.test.js`: `337 pass, 0 fail` --- _Generated by [Claude Code](https://claude.ai/code/session_01J4QvDfoDA4uk25z9qQLeWv)_ --------- Co-authored-by: Claude <noreply@anthropic.com>
…ipt (#115) ## Why Mathesis pins the standards proof-receipt reader at `e5a39d7`. That reader cannot read `(import …)` clauses, which resource-algebra's M-2 needs. The resource-algebra receipt that records M-2 now comes from hyperpolymath/resource-algebra#52, which supersedes #50. The reader on standards main (`940a27af`) can read those clauses. It also applies standards#1625 (owner ruling 2026-11-16): PWG-1 needs every counted checker to have run in a container image pinned by digest, named on the receipt. The owner chose to bump now and show the honest state. ## What **Reader pin.** `scripts/registry/claims-registry.mjs` sets `READER.commit` to `940a27af`, with the reader and checker digests updated (`proof-inputs-sha256.sh` is unchanged). `agda-proof-policy.yml` and `proven-tests.yml` fetch the same commit. **Mathesis's own receipt** (`verification/receipts/theory_praxis.deed`) is recorded again, with no new run. The new reader refused it for two reasons: - `:beholding-chora` used a `#u5` literal (owner ruling Q2). It now uses the canon form `#u8"chora:estate/github.com/hyperpolymath"`, the same value resource-algebra's receipt uses. - The proof lanes run in the pinned Guix shell, not a digest-pinned image, so the reader derives **PWG-2ii** for THY-1 and APP-1 to APP-4. The grades are recorded, and PROOF-STATUS and the receipt header give the reason. **Registry views** are rebuilt under the new reader. The new reader refuses every receipt that `claims.jsonl` pins, so **all 44 theorems show `ungraded`**, each with the reader's reason: | Repository | Pinned receipt | Why the reader refuses it | |---|---|---| | resource-algebra | at `47de4a8` | records PWG-1 where the reader derives PWG-2ii | | mathesis | at `58ebefb` | `#u5` `:beholding-chora` | | secret-types | | records PWG-1 where the reader derives PWG-2ii | | lintel | | `#u5` `:repo-uuid` | The mathesis pin is `58ebefb` since #114. **Theory ledger.** Main's #114 added entry 4. This PR appends entry 5: THY-1's change to `ungraded` under the new reader, as at the `e5a39d7` bump (`seq 2`). The append-only check against main passes. **Test fixture** gains an `image` option. PWG-1 holds with an image, and a new negative case shows that a recorded PWG-1 without one is refused (PWG-2ii). ## Still to come (follow-ups, not this PR) - **After resource-algebra#52 merges:** re-pin resource-algebra's rows to its receipt at `15fa113`, which gives PWG-2ii for 31 rows. Then add C-4b, E-1, E-2, E-3 (PWG-2ii) and M-2. M-2 is a Lean-only bridge with an `(import …)` of tropical-types, so it is ungraded. #52 waits on metadatastician/standards#1671. - **After this merges:** move mathesis's own five pins to the re-recorded receipt. - **secret-types and lintel:** their rows move when each home repository re-issues its receipt (a task is queued). ## Testing - **CI on `d26444a`:** every check passes, including the Guix proof lane "Agda, Rocq and Lean in the pinned Guix" and `CI/gate`. - **Locally,** with `MATHESIS_STANDARDS` at standards `940a27af`: - `claims-registry.mjs check` reports PASS: the views match, and the theory ledger is current and append-only (5 entries). - `bun test tests/registry`: 26 of 26 pass. - The census tests and the catalogue diff pass. - asciidoctor renders `PROOF-STATUS.adoc`. 🤖 Generated with [Claude Code](https://claude.com/claude-code) https://claude.ai/code/session_01LiQzWJdcPfRU6YMDGRG8Xi --------- Co-authored-by: Claude <noreply@anthropic.com>



Summary
The proof receipt is no longer written by hand.
scripts/proof-receipt.shis this repository's home script for the estate's shared receipt generator (standardsscripts/proof-receipt.sh). It writes the receipt with--writeand a newreceiptjob inproofs.ymlchecks it with--checkon every commit.The receipt describes
15fa113(the merge of #45) andproofs.ymlrun 44209774341, in which all three legs passed.The grades
Out of 32 theory rows the reader derives PWG-2ii for 31 and ungraded for 1 (M-2).
lake. That work is on its way as a separate change against main.Changes
scripts/proof-receipt.sh(new): the claim list, the run ledger and the hooks that name each leg's invocation and input set. It fetches the generator, the reader, the reference digest and the git checker at one pinned standards commit and checks each against its sha256.verification/receipts/resource-algebra-proofs_praxis.deed: generated at15fa113.scripts/check-lean-bridge.sh. That script holds the tropical-types pin and says of itself that a repin changes the home input digest; the hand-written manifests did not list it..github/workflows/proofs.yml: jobreceipt(pinned Bun by sha256, full-history checkout,--check), added to theProofs verdictjob'sneeds:.docs/status/PROOF-STATUS.adoc,docs/status/PROOF-NEEDS.adoc,CHANGELOG.adoc: the generator, the gate, the digests it computes, and what the gate asks of a later change.docs/status,CHANGELOG.adocand the receipt, resolved to the generated receipt and the documents that describe it.What the gate asks of a later change
A pull request that changes a leg's inputs (
src/, a recheck file, the bridge, the tropical-types pin) will fail thereceiptjob, because that leg's current receipt no longer describes the tree. The change marks the rowsupersededin the ledger and regenerates the receipt, so the grades fall to what the remaining receipts support (§8.4). A follow-up records the main run of the new commit. PROOF-STATUS says this where a reader meets it.Testing
2b9bf3f. The reader's output was identical for all 29 claims (after three generator fixes found by that comparison, which are in the standards PR).15fa113: same grades for all 32 claims; the differences are the checker strings and the extra Lean input.bash scripts/proof-receipt.sh --check: exit 0, endsPASS: negative control (the first claim recorded as :pwg PWG-1 is refused: its legs support PWG-2ii).--checkexit 1 with the stated reason, and the baseline passed before and after: an Agda input changed; the Rocq file changed; the tropical-types pin changed; a grade hand-edited to PWG-1; a row id removed from PROOF-STATUS.tests/workflows/ci_gate_test.sh: passes (the gate needs every other job).scripts/check-adoc-renders.sh: all 163 tracked.adocfiles render clean.check-no-md-in-docs.shandcheck-no-placeholders.shpass.tests/workflows/pass here. The other five need tools this container lacks or has in another form (nickel; ayqwith thekyamlformat;check_adoc_renders_test.shfails the same way on an untouched checkout of main). CI will show them.actionlintonproofs.yml: two SC2016 notices on a step this change does not touch; none from the new job.15fa113; this change touches none of their inputs, which the gate checks.RSR Quality Checklist
Required
.envfiles includedAs Applicable
CHANGELOGupdated🤖 Generated with Claude Code
https://claude.ai/code/session_013CuBToxr5h9catptGiTsFX