skills/aso-marketing/SKILL.md
Optimize App Store and Google Play listings with keyword strategy, metadata, and localization. Use when asked to improve app store visibility or ASO. Don't use for web SEO, paid UA campaigns, or pre-submission compliance audits.
npx skillsauth add luongnv89/skills aso-marketingInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
A comprehensive, iterative ASO workflow that takes a mobile app from analysis through planning, execution, verification, and reporting — covering both Apple App Store and Google Play Store.
Before running this skill, verify:
If any check fails, stop and ask for clarification.
Before creating, updating, or deleting files in an existing repository, sync the current branch with remote. Run the dry-run command first to preview, then confirm with the user before destructive sync.
# Dry-run preview first
git fetch --dry-run origin
git status
# Then sync (after user confirms working tree is clean or backed up)
branch="$(git rev-parse --abbrev-ref HEAD)"
git fetch origin
git pull --rebase origin "$branch"
If the working tree is not clean, ask the user before stashing. If origin is missing, pull is unavailable, or rebase/stash conflicts occur, stop and ask the user before continuing.
Backup-first rule for metadata files: Before overwriting any existing metadata file (e.g. metadata/version/1.0/en-US.json, fastlane/metadata/android/en-US/title.txt), create a .bak copy in the same directory or confirm the file is in version control. Never delete metadata files without explicit user confirmation.
This skill uses a Staged Pipeline + Review Loop architecture. Each phase maps to a subagent in agents/:
agents/analyzer.md — Phase 1 analysis reportagents/plan-writer.md — Phase 2 ASO planagents/compliance-checker.md — Phase 3 prohibited-keyword + trademark checkagents/executor.md — Phase 4 metadata changesagents/reviewer.md — Phases 5-6 review + best-practices verificationMain agent orchestrates phase transitions and the user approval gate after Phase 3.
The skill follows a 7-phase cycle. Never skip the policy compliance check or the planning approval gate — the user must approve a policy-compliant plan before execution begins.
Phase 1: Analyze → Phase 2: Plan → Phase 3: Policy Compliance Check
→ [User Approval Gate] → Phase 4: Execute → Phase 5: Review
→ Phase 6: Verify → Phase 7: Summarize
references/aso_best_practices.md for the complete ASO knowledge base — especially the "Store Policy Compliance" section detailing prohibited keywords, trademark rules, and listing policy restrictions for both Apple and Google.Run each phase in order. Detailed instructions, templates, and checklists live in the reference files. Read each one before executing the corresponding phase.
Read references/phases-1-2.md (Phase 1 section). Gather codebase info, audit existing store metadata, identify competitors, baseline performance, and produce an Analysis Report.
Read references/phases-1-2.md (Phase 2 section). Build a prioritized plan covering objectives, target keywords (primary/secondary/long-tail), metadata optimization (title, subtitle, keywords, description, what's-new), visual assets, localization, ratings strategy, and timeline.
Read references/phase-3-compliance.md. Validate every proposed metadata field against:
If any violation is found, revise the plan silently and re-run the compliance check. The user should only ever see a compliant plan.
After compliance passes, present the plan and compliance report to the user. Do not proceed to Phase 4 until the user explicitly approves the plan. Iterate as many times as needed; each revision must re-pass the compliance check.
Read references/phases-4-7.md (Phase 4 section). Implement approved changes systematically, in priority order. Before overwriting any existing metadata file, create a .bak backup or confirm version control coverage. Validate character limits after each edit.
Read references/phases-4-7.md (Phase 5 section). Run the review checklist on every changed field. Re-scan for prohibited keywords and trademark leaks.
Read references/phases-4-7.md (Phase 6 section) and cross-reference references/aso_best_practices.md. Final pass for policy compliance.
Read references/phases-4-7.md (Phase 7 section). Produce the final summary report (Changes Made, Metadata Comparison, Compliance Status, Expected Outcomes, Next Steps, Files Modified).
After each major step, output a status report with √ pass / × fail markers. Format and per-phase check lists are in references/edge-cases.md (Step Completion Reports section).
A complete run produces three artifacts: Analysis Report (Phase 1), ASO Plan + Compliance Report + updated metadata files (Phases 2-4), and Summary Report (Phase 7). Worked examples for each: see references/expected-output.md.
Input: an iOS+Android productivity app with weak metadata.
Title (iOS): FocusFlow: Work Timer (22/30, no primary keyword)
Keywords (iOS): timer,focus,work,productivity (34/100, 66 chars unused)
Short Desc (Android): A simple timer for focused work. (36/80, weak)
Output after Phase 4 (metadata/app-info/en-US.json):
{
"name": "FocusFlow: Focus & Work Timer",
"subtitle": "Deep Work Sessions & Tracking"
}
Output keywords field:
pomodoro,deep,work,concentration,study,block,distraction,habit,goal,flow,task
(97/100 chars, all prohibited terms cleared, no title/subtitle duplicates.) Full walkthrough in references/expected-output.md.
The skill run is successful when all metadata is within character limits, contains no prohibited keywords or trademarks, the user-approval gate was respected, and all required reports were produced. Full criteria list: references/edge-cases.md (Acceptance Criteria section).
Common edge cases (no local metadata files, single-platform app, no competitor access, pre-launch app, non-English primary locale, user rejects plan, compliance violation in draft) are documented in references/edge-cases.md. Read that file when an edge case applies — do not improvise.
Apple App Store, Google Play Store, and cross-store conventions (indexed fields, character limits, OCR-indexed screenshots, A/B testing surfaces) are documented in references/edge-cases.md (Platform-Specific Notes section).
This skill complements asc-aso-audit (deep iOS audit), asc-localize-metadata (bulk localization), asc-metadata-sync (App Store Connect sync), asc-whats-new-writer, asc-shots-pipeline, and seo-ai-optimizer. See references/edge-cases.md (Cross-Skill Integration section) for when to combine.
git pull --rebase, git stash, file deletion, or overwriting an existing metadata file, either run a dry-run preview, create a .bak backup, or ask the user. Failures and errors must surface to the user — do not silently retry.documentation
Manage software releases end-to-end: bump version, generate changelog, tag, push, GitHub release, publish to PyPI/npm. Use when user asks to ship, cut a release, tag a version, or list changes since last tag. Skip routine commits and marketplace publishing.
development
Review UI for usability issues using Steve Krug's principles and produce a scannable report. Use when asked for a usability audit, UX review, or UI feedback on screenshots, URLs, or code. Don't use for visual/brand design critique, accessibility (WCAG) audits, or backend/API review.
development
Validate app/startup ideas with market, feasibility, commercial, and open-source competitor analysis. Use when asked to evaluate, validate, or score a product idea. Don't use for PRDs, go-to-market plans, or investor decks.
testing
Install local-first security hardening: pre-commit secret detection, offline dependency scans, static analysis, reports, and gated free CI. Use when hardening repos or adding security hooks. Don't use for incident response or cloud security reviews.