llm: Add plan-android-work command and planning skills

[SaintPatrck]

🎟️ Tracking

Internal tooling improvement — Claude Code planning workflow infrastructure.

📔 Objective

Adds a pre-implementation planning workflow that pairs with /work-on-android. The new /plan-android-work command guides through requirements refinement and architecture planning before implementation begins.

New artifacts:

• plan-android-work command — 4-phase guided workflow (Ingest → Refine → Plan → Finalize) that takes Jira tickets, Confluence pages, or free-text descriptions and produces a design document saved to .claude/outputs/plans/
• refining-android-requirements skill — 5-category gap analysis rubric (Functional, Technical, Security, UX, Cross-cutting) covering product flavors, data layer tiers, streaming vs discrete data, sensitive ViewModel state, SDK crypto context isolation, and F-Droid compatibility. Includes blocking/non-blocking question classification and structured spec output (FR, TR, SR, UX numbered IDs)
• planning-android-implementation skill — 6-step architecture planning (Classify → Explore → Architecture → Files → Phases → Risk) with three-tier data layer diagram (Screen → ViewModel → Repository → Manager → DataSource), pattern anchors, file inventory guidance (Navigation, Module, Test files), phased task breakdown with co-located tests, and process death verification

CLAUDE.md update:

• Restructured Workflow Skills section into Planning Phase (skills 1-2) and Implementation Phase (skills 3-9)
• Updated quick start to reference both /plan-android-work and /work-on-android

[claude]

Claude finished @SaintPatrck's task in 6m 43s —— View job

---

Re-Review Complete

• Verified all 5 previously resolved threads
• Reviewed changes in commits 1cd8f9dd4 (Remove meta files) and f491cfd43 (Address review feedback)
• Full re-read of all 4 changed files in current state

Previous feedback addressed:

• Duplicate Execution Bridge removed from planning-android-implementation/SKILL.md -- the plan-android-work.md command now owns the sole "Executing This Plan" section in the final document template.
• MCP fallback guidance added to plan-android-work.md -- both a Prerequisites section and a "Tool unavailable" bullet in Source Detection Rules.
• Meta files (README, CHANGELOG, CONTRIBUTING) removed from both skill directories.

Current state assessment: Clean. The four files are internally consistent, the workflow phases are logically ordered, skill cross-references are correct, and the CLAUDE.md restructuring properly renumbers the skill list. The gap analysis rubric in refining-android-requirements and the phased planning structure in planning-android-implementation are comprehensive and well-tailored to Bitwarden Android patterns.

One minor note (not worth an inline comment): .claude/skills/refining-android-requirements/SKILL.md is missing a trailing newline -- the other two new files have one. Not blocking.

No new findings. This looks good to merge.

[github-actions]

Checkmarx One – Scan Summary & Details – 9b5b0520-c64b-43ec-a1a8-57714007ac3c

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

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 85.89%. Comparing base (d0dc420) to head (f491cfd).
⚠️ Report is 3 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #6597   +/-   ##
=======================================
  Coverage   85.89%   85.89%           
=======================================
  Files         803      803           
  Lines       57198    57198           
  Branches     8270     8270           
=======================================
  Hits        49130    49130           
  Misses       5195     5195           
  Partials     2873     2873           

Flag	Coverage Δ
app-data	17.54% <ø> (ø)
app-ui-auth-tools	20.94% <ø> (ø)
app-ui-platform	15.12% <ø> (ø)
app-ui-vault	25.87% <ø> (ø)
authenticator	6.45% <ø> (ø)
lib-core-network-bridge	4.29% <ø> (ø)
lib-data-ui	0.93% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ 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]

Thanks @david-livefront and @theMickster