agentleFS
batinbatin/five-guys/skills/agent-coordination/SKILL.md

agent-coordination

Skill2 starsChanged 51 days agoMITView on GitHub

5-agent (Architect/Data-Infra/Backend/Frontend/QA) sprint workflow, handoff protocol, board structure, and closing checklist. Used when starting a sprint/work item and during handoffs between agents.

13 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Files
SKILL.md

Agent Coordination Matrix

1. The Agent Team

AgentLayerOwns
architectARCHAPI Contract, Task Breakdown, ADR
backendBEAPI, Services, Business Rules
frontendFEUI, Forms, Dashboards
data-infraDB/INFRASchema, Migrations, Auth, CI
qaQAE2E Tests, Acceptance Testing, Security
<!-- SPRINT MODE START -->

2. Sprint Workflow (Required Order)

[Work Request] → Create Sprint Cards on the Board
                │
    1️⃣ architect: Prepare Contract
       • Data Model • API Endpoints • UI Flow • Error Codes
                │
        ┌───────┼────────┐
    2️⃣ data-infra   2️⃣ backend
    (Schema/Migration)  (API/Service) ──→ 2️⃣ frontend (Mock Mode/Screen)
        └───────┬────────┘
    3️⃣ qa: Acceptance Testing & Security
       ✅ Sprint Accepted or ❌ Back to Work

3. Handoff Protocol

FromToTagContent
architectdata-infra[SCHEMA-REQUEST]Table names, relationships, constraints
architectbackend[API-CONTRACT]Endpoint list, schema names, error codes
architectfrontend[UI-CONTRACT]Screen list, data requirements, mock schema
data-infrabackend[SCHEMA-READY]Migration ran, tables ready
data-infrafrontend[SEED-READY]Seed ran, deterministic IDs documented
frontend/backendqa[READY-FOR-QA]Sprint cards Done, ready for acceptance testing

Dispute: [CONTRACT-DISPUTE] card comment → architect decides and updates the contract.

4. Card Numbering Scheme

Format: [MODULE-LAYER-SEQ] — e.g. ORD-BE-003 = Orders module, Backend layer, 3rd card.

Layers: ARCH (architecture), DB (data), BE (backend), FE (frontend), INFRA (infrastructure), QA (quality).

5. Sprint Closing Checklist

Pre-Closing (when QA starts):

  • All ARCH cards Done ✅ (contract written)
  • All DB/BE/FE/INFRA cards Done ✅
  • QA card started

Post-QA (closing the sprint):

  • QA card Done ✅ (all scenarios green)
  • All bug cards triaged open/closed
  • Sprint retro held
<!-- SPRINT MODE END --><!-- NORMAL MODE START -->

2. Workflow (Required Order)

[Work Request] → Architect
                │
    1️⃣ architect: Prepare Contract
       • Data Model • API Endpoints • UI Flow • Error Codes
                │
        ┌───────┼────────┐
    2️⃣ data-infra   2️⃣ backend
    (Schema/Migration)  (API/Service) ──→ 2️⃣ frontend (Mock Mode/Screen)
        └───────┬────────┘
    3️⃣ qa: Acceptance Testing & Security
       ✅ Accepted or ❌ Back to Work

3. Handoff Protocol

Handoffs happen as direct notes/comments between agents (in code review, commit messages, or a shared doc) — no card IDs or board required.

FromToTagContent
architectdata-infra[SCHEMA-REQUEST]Table names, relationships, constraints
architectbackend[API-CONTRACT]Endpoint list, schema names, error codes
architectfrontend[UI-CONTRACT]Screen list, data requirements, mock schema
data-infrabackend[SCHEMA-READY]Migration ran, tables ready
data-infrafrontend[SEED-READY]Seed ran, deterministic IDs documented
frontend/backendqa[READY-FOR-QA]Work is done, ready for acceptance testing

Dispute: [CONTRACT-DISPUTE] note → architect decides and updates the contract.

4. Closing Checklist

Before QA starts:

  • Architecture contract written
  • DB/BE/FE/INFRA work complete

Before calling the work item done:

  • QA scenarios all green
  • All bugs found are triaged open/fixed
<!-- NORMAL MODE END -->

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.