Repository navigation
feat(receipt): write the receipt with the shared generator (standards f6ebc5e) - #54
Conversation
… f6ebc5e) scripts/grade-receipt.sh keeps this repository's ledgers and hooks and sources the standard's scripts/proof-receipt.sh (metadatastician/standards#1620), which writes and checks the receipt. The generator, the reader, the reference digest and the git checker are pinned at standards f6ebc5e, each by sha256, and fetched by --fetch-reader. - The receipt is unchanged but for the line naming the reader's commit; all six claims stay PWG-1. - The shared generator binds each Agda run row to the sha256 of the trusted base it checked (PROOF-POLICY 8.4). The current Agda row predates that field. Its value is the sha256 of proof_trusted_base in scripts/proof-flags.sh at the run's own commit 7dfeab2, the list that run's trusted-base gate checked the cone against; the list is unchanged since. The script says so beside the ledger. - proofs.yml: the proofs and legs jobs fetch the pinned tools before --run-row (every mode now sources the generator), and the Agda container installs curl to fetch them. - verification/legs/README.adoc no longer says the grade is PWG-2ii; the receipt derives PWG-1. Local: --write, then --check (git checks with the shared checker, the pinned reader at PWG-1 for N-1..N-3, N-7..N-9, the negative control) pass; --run-row for each leg gives the recorded inputs digest, and for Agda the same trusted-base binding. Not run here: the provers (CI runs them). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BYbQRtNu216R13jtZLmoqs
|
No actionable comments were generated in the recent review. Walkthrough
Merge Risk:
|
|
🟢 clean · commit Advisory: 11 row(s), worst 🔴 high (0 critical, 9 high) — reported, never grading the change.
Hypatia scan: 20 finding(s)
panic-attack assail: 16 finding(s)
pons asinorum: 9 finding(s)
finishingbot (release readiness): 174 finding(s)
glambot (presentation quality): 95 finding(s)
rhodibot (RSR compliance): 13 finding(s)
accessibilitybot: not_applicable — template default: enrol once the repository serves HTML
cipherbot: not_applicable — template default: enrol once the repository implements cryptography
echidnabot: not_applicable — template default: enrol once the repository carries proofs and an ECHIDNA engine is configured
panicbot: not_applicable — panic-attack is already graded by the verdict grid's own panic-attack row, from static-analysis-gate.yml evidence; enrolling panicbot would report one scanner twice and give two rows that can disagree
seambot: not_applicable — template default: enrol once a seam register exists at spec/seams/seam-register.json
sustainabot: not_applicable — reserved fleet slot; no adapter exists (gitbot-fleet bots/sustainabot/README.adoc)
Scope: static findings reported by each row's tool at the named commit. The table grades only findings in files this change touches, so the colour answers "does this change introduce anything?", not "what does the whole repository still carry?". Standing findings are counted in each accordion and in report.json; the baseline filter remains the place where acknowledged debt is recorded. The lane state is the worst GATING row; advisory rows are listed and counted but never set it. Purple, red, yellow and blue grade the worst GRADED finding present; orange is evidence missing, malformed or a scanner that did not run; grey is an explicit scope exclusion with its reason. No row is a security certification. report.json, badge.json and the raw findings are in the verdict-grid artifact on the linked run. |
check-trusted-base.sh calls grade-receipt.sh --inputs agda, which now sources the shared generator, so the fetch has to come before it. The proofs job failed at that step with "the pinned scripts/proof-receipt.sh is missing"; locally, with _build/standards removed, the same call fails the same way and passes once --fetch-reader has run. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BYbQRtNu216R13jtZLmoqs
|
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. |
|
You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard. |
…nned image (#57) ## Summary PROOF-POLICY §8.2 now counts a witness toward `PWG-1` only when its current receipt records the digest-pinned image its checker ran in. This comes from the owner ruling of 2026-11-16, landed as metadatastician/standards#1625. #54 pinned the receipt tools at standards `f6ebc5e`, which predates that rule. This PR moves the pin to `940a27a`, standards `main`, and does three things: - records what the last run can show; - states the grade the current reader derives; - runs the legs in a pinned image, so the next recorded run restores `PWG-1`. **Until that run is recorded, the committed receipt reads `PWG-2ii`.** The Rocq and Lean legs of run 43455458225 ran on the bare `ubuntu-24.04` runner. That runner has no digest, so under the current rule those witnesses cannot count toward `PWG-1`. ## Evidence from this PR's CI [Run 44214592597](https://github.com/hyperpolymath/secret-types/actions/runs/44214592597), on `1c28ec9`, ran all three legs in pinned images: - **Agda:** `debian:13-slim@sha256:a29215f6…`. - **Rocq and Lean:** `ubuntu:24.04@sha256:73b99939…`. - **Rows:** each row this run prints has the image as its 13th field. - **Grade:** the receipt of this run derives **`PWG-1`** for N-1..N-3 and N-7..N-9. - **Committed receipt:** passes `--check` at `PWG-2ii`. ## Changes - **Pins.** All four receipt tools are pinned at standards `940a27a`, each by sha256. - **Agda row.** It records `debian:13-slim@sha256:a29215f6…`, the container `proofs.yml` pinned for the Agda job at the run's own commit `7dfeab2`. - This is derived from that commit, in the same way #54 derived the trusted-base binding. - The Rocq and Lean rows carry no image, because none existed. - **`proofs.yml`.** - The legs job runs in `ubuntu:24.04@sha256:73b99939…`, the image lintel#31 uses. It installs its tools as root and marks the workspace safe for git, as the Agda job does. - Both jobs pass `--image` to `--run-row`. The value is read from the job's own `container:` line at the commit under test, so there is one source of truth. - Both jobs also pass `--axioms none`, which the generator at `940a27a` requires. The gates refuse any axiom. - **Docs.** README, PROOF-STATUS, the legs README, the proof-debt ledger and the CHANGELOG now say `PWG-2ii`, and why. ## After merge Record the first run on `main`, using the rows its receipt job prints. All three legs then carry an image, and the reader derives `PWG-1` again. ## RSR Quality Checklist ### Required - [x] Tests pass. CI is green on `1c28ec9`, including the Agda, legs and proof-receipt jobs. Locally, `--write` and `--check` pass. - [x] No banned language patterns (bash, YAML and AsciiDoc only). - [x] No banned functions. - [x] SPDX headers unchanged. - [x] No secrets. ### As Applicable - [x] Documentation updated. - [x] `CHANGELOG` updated. ## Testing ``` PASS: verification/receipts/agda_praxis.deed matches the tree and the ledgers (source-commit 292d566) PASS: git checks of §8.5.3 (shared checker, standards 940a27af1e25, --branch refs/remotes/origin/main) PASS: the pinned reader (standards 940a27af1e25) accepts verification/receipts/agda_praxis.deed N-1 PWG-2ii foundation-diverse false ... PASS: negative control (the first claim recorded as :pwg PWG-1 is refused: its legs support PWG-2ii) ``` 🤖 Generated with [Claude Code](https://claude.com/claude-code) https://claude.ai/code/session_01BYbQRtNu216R13jtZLmoqs --------- Co-authored-by: Claude <noreply@anthropic.com>



Summary
The proof receipt is now written by the standard's shared generator,
scripts/proof-receipt.shfrom metadatastician/standards#1620. That generator was lifted from this repository's ownscripts/grade-receipt.sh. The receipt is unchanged except for the line naming the reader's commit, and all six claims stayPWG-1.Changes
scripts/grade-receipt.shkeeps this repository's parts: the claim, run and echo ledgers, and the hooks that name each leg's invocation and input set. It then sources the generator.f6ebc5e, each by sha256: the generator, the reader, the reference digest and the git checker.--fetch-readerfetches all four.proof_trusted_baseinscripts/proof-flags.shat the run's own commit7dfeab2.0c1f64c3….proofs.yml:proofsandlegsjobs fetch the pinned tools before they use them, because every mode now sources the generator.proofsjob the fetch comes before the trusted-base gate, which callsgrade-receipt.sh --inputs agda.curlto fetch them.verification/legs/README.adocno longer says the grade isPWG-2ii, which contradicted the receipt's derivedPWG-1.CHANGELOG.adocentry.RSR Quality Checklist
Required
e861780, and the proofs, legs and proof-receipt jobs all pass.PWG-1for N-1..N-3 and N-7..N-9.--check: the tree and ledgers match, the §8.5.3 git checks pass with the shared checker, the pinned reader accepts it, and the negative control holds.As Applicable
verification/legs/README.adoc).CHANGELOGupdated.Testing
🤖 Generated with Claude Code
https://claude.ai/code/session_01BYbQRtNu216R13jtZLmoqs
Summary by CodeRabbit
Documentation
Chores