Repository navigation
fix(receipt): record PWG-2ii for N-1, N-2, N-3, N-7, N-8, N-9 (standards#1625) - #55
hyperpolymath wants to merge 3 commits into
Conversation
…rds#1625) standards#1625 (owner ruling 2026-11-16) makes PWG-1 require that every counted witness's current receipt names the container image its checker ran in, pinned by digest (:image "NAME@sha256:<64 hex>"). None of this receipt's legs carries an :image, so the reader at standards 940a27af derives PWG-2ii. Set :pwg PWG-2ii on all six claims and record the downgrade in PROOF-STATUS.adoc. No digest was added. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01J4QvDhq2buNFimmickMwwp
|
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: 17 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. |
…standards 940a27a The generator still graded PWG-1 whenever three witnesses described the source, so --check failed against the PWG-2ii receipt and --write would revert it. pwg_of now mirrors readProofReceipt at standards 940a27af: PWG-1 also needs every counted witness's current receipt to name a digest-pinned :image. The run ledger gains an optional 11th image column, emitted as :image; no row carries one, as no CI log printed a digest. Pin the reader, reference digest script and checker at standards 940a27af, regenerate the receipt with --write (only the header pin changes), and correct the stale PWG-1 statements in README.adoc, docs/proof-debt.adoc and verification/proofs/README.adoc. Add a CHANGELOG entry. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01J4QvDhq2buNFimmickMwwp
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01J4QvDhq2buNFimmickMwwp
|
|
Closing in favour of #56, at the owner's choice. #54 merged the shared generator into I checked #56's receipt at head The mathesis Generated by Claude Code |



Summary
The standards proof-receipt reader (standards 940a27af) rejected
verification/receipts/agda_praxis.deed:FAIL ...: claim "N-1" records :pwg PWG-1, but its legs and receipts support PWG-2ii (PROOF-POLICY §8.2, §8.4)Cause: standards#1625 (owner ruling 2026-11-16). PWG-1 now requires every counted witness's current receipt to name the container image its checker ran in, pinned by digest (
:image "NAME@sha256:<64 hex>"). None of this receipt's legs names one, so the reader derives PWG-2ii. No digest was added or invented: no CI log of these runs printed one. The Agda job runs in a digest-pinned container, but the Rocq and Lean legs run on the bare runner, so PWG-1 would not be reachable by naming an image on the Agda leg alone.The fix is made at its source, the generator
scripts/grade-receipt.sh, so that CI's--checkpasses and a later--writedoes not revert the grades.Changes
scripts/grade-receipt.sh:pwg_ofnow mirrorsreadProofReceiptat standards 940a27af. PWG-1 needs three source-describing witnesses and a current receipt naming a digest-pinned:imageon every counted witness. Otherwise it is PWG-2ii.imagecolumn, emitted as:imageon a current receipt. No row carries one.STANDARDS_COMMITis now 940a27af1e25a73a2d55daa78d0763e6749fd935, with the reader and checker sha256 pins updated (the digest script is unchanged).verification/receipts/agda_praxis.deed: six:pwgfields, plus the header's reader pin (now 940a27a).--writereproduces the file exactly, so it is the generator's own output.PROOF-STATUS.adoc: the census records PWG-2ii (was PWG-1) with the reason, and names the 940a27a pin.README.adoc,docs/proof-debt.adoc,verification/proofs/README.adoc: stale PWG-1 statements corrected to PWG-2ii with the #1625 reason.CHANGELOG.adoc: new entry.New receipt sha256:
2c5f13fe59bfe49d719ba70068f4a63fc0215d2f022773b4d72ca9420c8138acCommits on the branch:
d8cb4fea3fb5fef59b539e27553744e6ce3532a7(receipt and census),56c525998c15ae256e682e2ea791302f8be7c724(generator, pin, docs),2be7d3e7ae04133de2e9a8e6101e7c164b31d07e(PROOF-STATUS pin text). The head is2be7d3e.Testing
All at standards 940a27af.
node 1-formats/deed/tools/deed_lint.js --proof-receipt-json verification/receipts/agda_praxis.deed: exit 0. It derives PWG-2ii for N-1, N-2, N-3, N-7, N-8 and N-9.scripts/check-proof-receipt.sh <receipt> <full clone>: exit 0. It prints 28oklines and ends withcheck-proof-receipt: every check holds.bash scripts/grade-receipt.sh --check: exit 0:scripts/check-adoc-renders.sh(asciidoctor is not installed here).%G?showsNbecause no allowed-signers file is configured, so verify them on the forge.Follow-up
docs/atlas/claims/claims.jsonlrows for secret-types currently pin this receipt at commit 4ed0672 / sha256 4ed7e550e399074cc8a93bd190f193a0fea78de97d54e27889263b9d386ce9cc. They need re-pinning to the merge commit and the new sha256 (2c5f13fe59bfe49d719ba70068f4a63fc0215d2f022773b4d72ca9420c8138ac, if the receipt does not change before merge) once this merges (and lintel's equivalent PR merges).🤖 Generated with Claude Code
https://claude.ai/code/session_01J4QvDhq2buNFimmickMwwp
Summary by CodeRabbit