Every feature request or bug report enters through the Captain. Specialist bots work in parallel where it is safe, share one cache instead of re-reading raw documents, and stop at six human approval gates. Click any bot to see what it does.
| # | Agent | Does | Writes | Next |
|---|---|---|---|---|
| 1 | Captain | Reads the input (task key, BRS, design, request) and classifies it | Task folder | A "?" gets an answer only. Otherwise Intake starts |
| 2 | Intake analysts | BRS, tracker, design and code read once, in parallel | brief, tracker, design, code .md | The owner approves, then Gap Finder starts |
| 3 | Gap Finder | Finds gaps and blocker questions | gap.md | The owner answers, then Dev Planner starts |
| 4 | Dev Planner + Task Drafter | Plan, test cases and ticket drafts | plan.md, tracker.md | The owner approves, or answers go back to the planner |
| 5 | Git Bot | Repo-ready check (advice only) | Note | The owner readies the branch, then devs start |
| 6 | Flutter Devs + QA Test Writer | Code plus tests, and QA cases | Code, qa.md cases | Analyzer gate |
| 7 | Analyzer gate | fvm flutter analyze and related tests | Result | Pass goes to Review. Fail goes back to devs |
| 8 | Review bots | PR Reviewer, Code Quality, Security and Localization, in parallel | review.md, quality.md | The owner picks fixes. Comments go back to devs (loop to 6) |
| 9 | QA Lead + testers | Multi-language tests on devices | qa/ evidence | Fail means rework (loop to 6). Pass goes to the report |
| 10 | QA Reporter | Fills the report template (HTML or PDF) | Final report | The owner accepts the verdict |
| 11 | Tracker Updater | Posts the comment and status | Tracker | The owner says yes, then tracker Done |
| 12 | Cache Keeper | Saves repeat issues and syncs rules | rules, must-haves | Next task starts smarter |
flutter-captain, flutter-clean-feature, flutter-legacy-edit, flutter-responsive-ui, flutter-reusable-widget, flutter-feature-review, <money-feature-module>, coworker-engineer)<AI-ROOT>/<App-Name>/tasks/<TASK-KEY>/<HOME>/.cursor/agents/<prefix>-*.md/<prefix>-captain <TASK-KEY> inside Cursor on the repo<HOME>/.cursor/skills and <REPO>/.cursor/skills<REPO>/.cursor/rules/*.mdc<REPO>/.cursor/agents, skills, rules and AGENTS.md| File | What | Why |
|---|---|---|
PROJECT_CONTEXT.md | App overview | Read once |
project/rules.md | Team rules | Single source |
project/MEMORY.md | Dated decisions | No repeat discussions |
project/quality-rules.md | Code quality rules | Local quality gate |
project/quality-must-haves.md | Pre-dev checklist | Right on the first pass |
project/devices.md | QA devices | Same devices every run |
project/app-internals.md | App internals | No re-deriving |
project/test-accounts.md | Test logins | Local only, never shared |
tasks/<KEY>/ brief, gap, plan, quality, review, qa .md | Stage summaries | Each stage reads a short file to save tokens |
qa/ | Screenshots, videos, reports | QA evidence |
_shared/qa-kit/ | Tools, report template | Reused by every app |
๐ Cache Keeper keeps both sides in sync, so Grok and Cursor follow the same rules and files.
Transfers, wallet, payments, history
<money-feature-module>Plans, bundles, balance, profile, remote config
Shared by both: approval gates, fvm only, light TDD, code quality grade B+ with 80% coverage on changed code, pixel QA against the design, and git done only by the owner.
<AI-ROOT>/<App-Name>/PROJECT_CONTEXT.md<AI-ROOT>/<App-Name>/project/ rules.md, MEMORY.md, test accounts (local only)<AI-ROOT>/<App-Name>/tasks/<TASK-KEY>/ stage .md files<AI-ROOT>/_shared/qa-kit/report-template/<REPO>/.cursor/ and ~/.cursor/agentsBased on the team's own playbooks (team-builder skill and pipeline docs). Optional bots such as the Crash, Analytics and Localization checks can be switched off per project. This page shows how work flows. It contains no performance figures.