Skip to content

[PM-33575] llm: Add delivery skills for preflight, commit, PR, and labeling#2449

Merged
SaintPatrck merged 4 commits intomainfrom
llm/delivery-skills
Apr 8, 2026
Merged

[PM-33575] llm: Add delivery skills for preflight, commit, PR, and labeling#2449
SaintPatrck merged 4 commits intomainfrom
llm/delivery-skills

Conversation

@SaintPatrck
Copy link
Copy Markdown
Contributor

🎟️ Tracking

https://bitwarden.atlassian.net/browse/PM-33575

📔 Objective

Adds four Claude skills covering the delivery phase of iOS feature development: pre-commit quality gate, commit message format, PR creation, and PR labeling.

What changed:

  • skills/perform-ios-preflight-checklist — Automated checks (swiftformat, swiftlint, typos) followed by a manual checklist covering architecture (UDF, domain constraints), security (zero-knowledge, Keychain, InputValidator), testing (action/effect tests, error paths, AutoMockable), documentation/style (DocC, JIRA TODOs, naming), and hygiene (no secrets, no force-unwraps, no accidental spec changes).

  • skills/committing-ios-changes[PM-XXXXX] <type>: description format with type guide and real examples from git log. Guidance on staging specific files and what not to commit (credentials, build artifacts, unintended spec changes, snapshot images).

  • skills/creating-ios-pull-request — PR title format matching commit convention, body template from .github/PULL_REQUEST_TEMPLATE.md (Tracking / Objective / Screenshots), always create as draft, post-creation steps (label, self-review, mark ready).

  • skills/labeling-ios-changes — Change type labels (t:feature-app, t:feature-tool, t:bug, t:tech-debt, t:docs, t:ci, t:deps, t:misc, t:breaking-change), app context labels (app:password-manager, app:authenticator), and special labels (ai-review, automated-pr, hold, needs-qa).

…beling

- perform-ios-preflight-checklist: automated checks (swiftformat, swiftlint,
  typos) + manual checklist covering architecture, security, testing,
  documentation/style, and hygiene
- committing-ios-changes: [PM-XXXXX] <type>: format with type guide and
  examples from git log; staged file guidance; what not to commit
- creating-ios-pull-request: PR title format, body template from
  .github/PULL_REQUEST_TEMPLATE.md (Tracking/Objective/Screenshots),
  always draft, post-creation steps
- labeling-ios-changes: change type labels (t:feature-app, t:bug,
  t:tech-debt, etc.), app context labels (app:password-manager,
  app:authenticator), special labels (ai-review, hold, needs-qa)
@github-actions github-actions bot added the t:llm Change Type - LLM related change (e.g. CLAUDE.md files) label Mar 13, 2026
@github-actions
Copy link
Copy Markdown
Contributor

github-actions bot commented Mar 13, 2026

Logo
Checkmarx One – Scan Summary & Details5bf34826-0d7e-4c6e-b3e9-e97f15c545ff

Great job! No new security vulnerabilities introduced in this pull request

@SaintPatrck SaintPatrck added the ai-review Request a Claude code review label Mar 13, 2026
@claude
Copy link
Copy Markdown

claude bot commented Mar 13, 2026

Claude finished @SaintPatrck's task in 2m 4s —— View job


Code Review Summary

Reviewed all 4 skill files added in this PR. The skills are well-structured, consistent with project conventions, and provide clear guidance for the delivery workflow.

Two minor findings were posted as inline comments:

  1. committing-ios-changes/SKILL.md — The commit type guide table omits the llm type, which is actively used in this repo (including by this very PR: [PM-33575] llm: ...).
  2. labeling-ios-changes/SKILL.md — The change type label mapping omits t:llm, which is an existing label applied to this PR itself.

Both are completeness gaps — the skills should document the patterns they themselves follow.

Otherwise, the PR looks good. The four skills cover the delivery workflow clearly with appropriate references to project conventions, tooling, and CLAUDE.md guidelines.


  • Gather context and read changed files
  • Review skill content for correctness and completeness
  • Post inline findings
  • Post summary

@codecov
Copy link
Copy Markdown

codecov bot commented Mar 13, 2026

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 85.81%. Comparing base (48115b8) to head (ca56d9a).
⚠️ Report is 58 commits behind head on main.

Additional details and impacted files
@@            Coverage Diff             @@
##             main    #2449      +/-   ##
==========================================
- Coverage   86.86%   85.81%   -1.06%     
==========================================
  Files        1841     2082     +241     
  Lines      162244   178396   +16152     
==========================================
+ Hits       140941   153085   +12144     
- Misses      21303    25311    +4008     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

@SaintPatrck SaintPatrck marked this pull request as ready for review March 27, 2026 18:32
@SaintPatrck SaintPatrck requested review from a team as code owners March 27, 2026 18:32
@SaintPatrck SaintPatrck added ai-review-vnext Request a Claude code review using the vNext workflow and removed ai-review Request a Claude code review labels Mar 27, 2026
@github-actions
Copy link
Copy Markdown
Contributor

github-actions bot commented Mar 27, 2026

Bitwarden Claude Code Review

Overall Assessment: APPROVE

Reviewed all 5 changed files: the CLAUDE.md delivery workflow section and 4 new skill files (preflight checklist, commit formatting, PR creation, PR labeling). The skills are well-structured, internally consistent, and accurately reflect the conventions documented in CLAUDE.md. The two findings from the prior review (missing llm commit type and t:llm label) have been addressed in the latest revision. No new issues identified.

Code Review Details

No findings to report. All four skills correctly reference existing tooling (swiftformat, swiftlint, typos), follow the repository's commit and PR conventions, and cross-reference each other appropriately (e.g., preflight before commit, labeling after PR creation).

@SaintPatrck SaintPatrck merged commit 8bf3557 into main Apr 8, 2026
20 checks passed
@SaintPatrck SaintPatrck deleted the llm/delivery-skills branch April 8, 2026 13:00
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ai-review-vnext Request a Claude code review using the vNext workflow t:llm Change Type - LLM related change (e.g. CLAUDE.md files)

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants