From 76118942d4b2a500fcff348555b517fbf8a44fe7 Mon Sep 17 00:00:00 2001 From: ItzikEzra-rh Date: Tue, 9 Jun 2026 11:19:47 +0300 Subject: [PATCH 1/2] Add supply chain security checks to review protocol --- _shared/review-protocol.md | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/_shared/review-protocol.md b/_shared/review-protocol.md index e775be8..7d3e746 100644 --- a/_shared/review-protocol.md +++ b/_shared/review-protocol.md @@ -41,7 +41,11 @@ Evaluate changes against these categories, prioritized by impact: 2. **Error handling** — Are errors caught, propagated, and reported appropriately? Are failure modes handled? 3. **Security** — Are there injection risks, unsafe operations, exposed secrets, - or other OWASP-category concerns? + or other OWASP-category concerns? Reject any PR introducing `.claude/` or + `.vscode/` directories (active supply chain attack vectors). Flag changes to + CI/CD configuration (Actions workflows, pipeline configs) for extra scrutiny. + If the most recent commit author differs from the PR author, treat as + potentially compromised. 4. **Design** — Does each new abstraction earn its complexity? Are responsibilities clearly divided — no god functions, no single type accumulating unrelated concerns? Do interfaces hide implementation details From cc381d8c4bf232b8c82e34c68fa9c66909c5fdc2 Mon Sep 17 00:00:00 2001 From: ItzikEzra-rh Date: Tue, 9 Jun 2026 13:38:32 +0300 Subject: [PATCH 2/2] Address review: improve actionability of author mismatch guidance --- _shared/review-protocol.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/_shared/review-protocol.md b/_shared/review-protocol.md index 7d3e746..51ed47c 100644 --- a/_shared/review-protocol.md +++ b/_shared/review-protocol.md @@ -44,8 +44,8 @@ Evaluate changes against these categories, prioritized by impact: or other OWASP-category concerns? Reject any PR introducing `.claude/` or `.vscode/` directories (active supply chain attack vectors). Flag changes to CI/CD configuration (Actions workflows, pipeline configs) for extra scrutiny. - If the most recent commit author differs from the PR author, treat as - potentially compromised. + If the most recent commit author differs from the PR author, flag for + manual review before approving. 4. **Design** — Does each new abstraction earn its complexity? Are responsibilities clearly divided — no god functions, no single type accumulating unrelated concerns? Do interfaces hide implementation details