Repository navigation
chore: Re-apply the repo-ci-baseline Template - #332
Conversation
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
…rom the Template Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #332 +/- ##
=======================================
Coverage 85.42% 85.42%
=======================================
Files 117 117
Lines 2717 2717
Branches 387 387
=======================================
Hits 2321 2321
Misses 318 318
Partials 78 78 🚀 New features to boost your workflow:
|
There was a problem hiding this comment.
Claude Review
Overview
This PR does two things:
- It narrows the long-token redaction rule in
post_claude_review.pyand its inline copy inclaude-review.yml, so only long tokens that contain a credential start are replaced. - It adds an opt-in
ui(UI developer) Sandcastle role, with fixtures and tests.
The redaction change is sound. Every CREDENTIAL alternative needs one of the listed starts inside a single token chain, so no credential escapes, and the scan stays linear. The ui wiring (uiEnabled, resolveRoles, buildIssue, backendPromptArgs) also looks correct. I found one blocking-worthy problem: the shipped roles/ui.md has no # SKILLS section, so the new skills/prompt consistency test fails for any repo that opts in.
Changes
| File | What changed |
|---|---|
.github/scripts/post_claude_review.py, .github/workflows/claude-review.yml |
CREDENTIAL_START added. The long-run replacement now requires a credential start in the chain. The workflow copy is kept in sync. |
.github/scripts/tests/test_post_claude_review.py |
Tests for long plain tokens left alone, long credential starts still replaced, script and workflow parity, and every CREDENTIAL match containing a start. |
.sandcastle/lib/agents.mts, config.mts, plan.mts, build.mts, prompts.mts |
AnyRole/OptInRole, uiEnabled, roleAgent, and ui in OPTIONAL_ROLES and BUILD_ROLES. resolveRoles drops ui when not enabled. buildIssue runs ui after the backend. backendPromptArgs sets UI_DEVELOPER. |
.sandcastle/roles/backend.md, roles/ui.md |
backend.md uses {{UI_DEVELOPER}}. ui.md is a new prompt. |
.sandcastle/lib/ui-role-fixture.mts and tests (agents, build, config, plan, skills, ui-role-fixture) |
withUiRole fixture, ui build tests, a prompt/skills consistency test, and a config check. |
Many *.test.mts, git-sandbox-fake.mts |
rmSync now retries (maxRetries: 5, retryDelay: 100). |
.sandcastle/main.mts |
Header comment updated. |
Advisory (not blocking)
- The
uirole's prompt-existence checks inconfig.test.mtsandskills.test.mtsare skipped in this repo, becauseuiis not opted in. Nothing in CI exercises the opt-in path against the realui.md. That is how the finding went unnoticed. withUiRolemutates the sharedROLE_AGENTSobject, anddelete agents.uiwould throw in strict ESM ifROLE_AGENTSwere ever frozen.backend.mdnow requires{{UI_DEVELOPER}}unconditionally. I could not confirm thatbackendPromptArgsis the only thing that rendersbackend.md. If anything else renders it, Sandcastle rejects the missing placeholder.ui.mdsays little about what to do when a tester's test is wrong rather than the code, beyond "leave it and say why in your commit body".
Not reviewed
I did not read AGENTS.md, CLAUDE.md, repo-config.mts, plan-prompt.md, or the host() helper in build.test.mts. Statements about them are inferred from the diff and the context pieces.
Automated release blog posts for #332, opened by the release workflow. It holds every Release whose post isn't on main yet, rebuilt from main on each run. The [skip-release] title marker keeps its merge from starting another release. Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Why
Re-Apply the repo-ci-baseline Template so this repo carries the Template changes from mpaulosky/dotfiles#303, #304, #305 and #306: the opt-in UI developer role, the narrowed long-run credential redaction in Claude Review's script, the shared ui-role test helper and the retried scratch-directory cleanup in the Sandcastle tests.
What changed
2a179ebApply commit,apply.shoutput alone: 22 Owned files (byte-identical to the Template's) and one new Seed file,.sandcastle/roles/ui.md, the UI developer's prompt, which stays unused here: this repo has nouientry inROLE_AGENTS.17e8da3Takes the UI developer line in.sandcastle/roles/backend.mdfrom the Seed: the Ownedlib/prompts.mtsalways passesUI_DEVELOPER, so a backend prompt without the placeholder logged an unreferenced-argument warning on every run. With nouientry the line renders the sentence it had before.Verification
scripts/gate.sh) passed./pre-push-review: 1 round, no findings.🤖 Generated with Claude Code