Guidance audit & public showcase: 38-finding audit, zero-loss streamline, Cavalry front door - #4
Closed
DeyangChan wants to merge 12 commits into
Closed
Guidance audit & public showcase: 38-finding audit, zero-loss streamline, Cavalry front door#4DeyangChan wants to merge 12 commits into
DeyangChan wants to merge 12 commits into
Conversation
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…logs A fork PR from a branch named 'main' with green CI previously satisfied the workflow_run condition and would deploy the fork's commit with repo secrets. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Root 2488->1953w, backend 2817->2228w, db 560->438w per the audit ledger; single-owner rules with pointers replace cross-file restatements; db gains the hoisted money/timestamps/unique-constraint rules. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…P hardening specs/README admits spec-tool directories; design/README owns the mockup loop; otp-auth gains CSPRNG/single-use/bounded-TTL/max-attempts rules (F-005). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
README rewritten as the public front door with theme-aware Cavalry lockup; design guide carries the mark and attribution from local assets only. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
README states the JS/TS home turf, enumerates packs and add-ons, explains contract auto-loading, links Cavalry, and de-vendors the staging step. stacks/README.md now mandates five files with n/a stubs, a shared per-area skeleton incl. security bindings, replacements-only registers, standalone packs. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Frontend 5051->3643w (single-owner tier/floor/check rules, gates as the one normative digest, two rules hoisted from packs); infra 1778->1371w (GCP as replaceable default binding, Guardrails owns approval/environment rules once). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…acking Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
nextjs 7748->5927w (+infra.md n/a stub, security/add-on bindings, registers cut to replacements); taro 5545->4596w (self-contained, generalized values); vercel 6320->5047w (+develop-branch and analytics register entries). All ledger rules kept, merged with pointers, or hoisted to base owners. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Contributor
Author
|
Closing as superseded. This branch predates the template overhaul that landed in #10 and now in #11. Its stack packs are the retired Reviewed for salvage before closing. Everything here is either already in
Deleting the branch. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Spec:
specs/002-audit-repo-guidance/(spec.md · plan.md · tasks.md — 41/44 tasks complete; 3 gated on maintainer decisions below)What this PR does
The P1 deep audit of the whole repo, plus every non-destructive fix it authorized — four workstreams:
specs/002-audit-repo-guidance/audit-report.md: 38 findings, severity-ranked, 100% file coverage, instruction-discovery map answering the "does Claude miss README.md files?" question (yes it would — wiring now guarantees ≤1 hop from auto-loaded guidance), gitleaks-clean history, zero broken links.rule-inventory.md). Per-session agent context (root + area + pack appendix) down ~25%. Highlights: frontend contract 5,051→3,643; root 2,488→1,953; every rule single-owner with pointers.LICENSE, design-guide attribution, all brand assets local. Fresh-reader test ×2: "reads convincingly as the public flagship of an opinionated senior team."infra.mdn/a-stub); conflict registers hold replacements only; the two silent base contradictions (verceldevelopbranch, pinned analytics) now registered exceptions; security + add-on bindings added where the base delegated them; zero orphaned instruction files.Security fixes worth review attention:
deploy.ymlfork-PR gate (event == 'push'guard) · otp-auth hardening (CSPRNG, single-use, bounded TTL, max attempts).Test plan
gitleaks git . --redact— 54 commits, no leaks (scan-evidence.md)lint / typecheck / test / build— cannot run: the toolchain is the template's intentional<pm>/TODO placeholder (stated per Definition of Done)Decisions requested before merge (audit-report.md § Decisions requested)
.claude/,.specify/— untracked and unignored): recommended gitignore.specs/001-*,specs/002-*): recommended not shipped in the public template.🤖 Generated with Claude Code