Team Interactive Craft

Team Interactive Craft

Systematic interactive component pipeline: research -> interaction design -> build -> a11y test. Built on team-worker agent architecture -- all worker roles share a single agent definition with role-specific Phase 2-4 loaded from roles/<role>/role.md.

Architecture

Skill(skill="team-interactive-craft", args="task description")
                    |
         SKILL.md (this file) = Router
                    |
     +--------------+--------------+
     |                             |
  no --role flag              --role <name>
     |                             |
  Coordinator                  Worker
  roles/coordinator/role.md    roles/<name>/role.md
     |
     +-- analyze -> dispatch -> spawn workers -> STOP
                                    |
                    +-------+-------+-------+-------+
                    v       v       v       v
           [team-worker agents, each loads roles/<role>/role.md]
        researcher  interaction-designer  builder  a11y-tester

Role Registry

Role Router

Parse $ARGUMENTS:

• Has --role <name> -> Read roles/<name>/role.md, execute Phase 2-4
• No --role -> roles/coordinator/role.md, execute entry router

Delegation Lock

Coordinator is a PURE ORCHESTRATOR. It coordinates, it does NOT do.

Before calling ANY tool, apply this check:

If a tool call is BLOCKED: STOP. Create a task, spawn a worker.

No exceptions for "simple" tasks. Even a single-file read-and-report MUST go through spawn_agent.

Shared Constants

• Session prefix: IC
• Session path: .workflow/.team/IC-<date>-<slug>/
• CLI tools: ccw cli --mode analysis (read-only), ccw cli --mode write (modifications)
• Message bus: mcp__ccw-tools__team_msg(session_id=<session-id>, ...)
• Max GC rounds: 2

Worker Spawn Template

Coordinator spawns workers using this template:

spawn_agent({
  agent_type: "team_worker",
  task_name: "<task-id>",
  fork_turns: "none",
  message: `## Role Assignment
role: <role>
role_spec: <skill_root>/roles/<role>/role.md
session: <session-folder>
session_id: <session-id>
requirement: <task-description>
inner_loop: <true|false>

Read role_spec file (<skill_root>/roles/<role>/role.md) to load Phase 2-4 domain instructions.

## Task Context
task_id: <task-id>
title: <task-title>
description: <task-description>
pipeline_phase: <pipeline-phase>

## Upstream Context
<prev_context>`
})

After spawning, use wait_agent({ timeout_ms: 1800000 }) to collect results. If result.timed_out, send STATUS_CHECK via followup_task (wait 3 min), then FINALIZE with interrupt (wait 3 min), then mark timed_out and close agents. Use close_agent({ target }) each worker.

Model Selection Guide

Interactive craft is a technical pipeline where research informs interaction design, which guides implementation. Builder needs creative problem-solving for vanilla JS constraints, a11y-tester needs thorough analysis.

Research-to-Design Context Flow

Researcher findings must reach interaction-designer via coordinator's upstream context:

// After RESEARCH-001 completes, coordinator sends findings to interaction-designer
spawn_agent({
  agent_type: "team_worker",
  task_name: "INTERACT-001",
  fork_turns: "none",
  message: `## Upstream Context
Research findings: <session>/research/interaction-inventory.json
Browser API audit: <session>/research/browser-api-audit.json
Pattern reference: <session>/research/pattern-reference.json`
})

User Commands

Specs Reference

• specs/pipelines.md -- Pipeline definitions and task registry
• specs/interaction-patterns.md -- Interaction pattern catalog
• specs/vanilla-constraints.md -- Zero-dependency rules

Session Directory

.workflow/.team/IC-<date>-<slug>/
+-- .msg/
|   +-- messages.jsonl         # Team message bus
|   +-- meta.json              # Pipeline config + GC state
+-- research/                  # Researcher output
|   +-- interaction-inventory.json
|   +-- browser-api-audit.json
|   +-- pattern-reference.json
+-- interaction/               # Interaction designer output
|   +-- blueprints/
|       +-- {component-name}.md
+-- build/                     # Builder output
|   +-- components/
|       +-- {name}.js
|       +-- {name}.css
+-- a11y/                      # A11y tester output
|   +-- a11y-audit-{NNN}.md
+-- wisdom/                    # Cross-task knowledge

v4 Agent Coordination

Message Semantics

Agent Health Check

Use list_agents({}) in handleResume and handleComplete:

// Reconcile session state with actual running agents
const running = list_agents({})
// Compare with tasks.json active_agents
// Reset orphaned tasks (in_progress but agent gone) to pending

Named Agent Targeting

Workers are spawned with task_name: "<task-id>" enabling direct addressing:

• send_message({ target: "INTERACT-001", message: "..." }) -- send research findings to interaction-designer
• followup_task({ target: "BUILD-001", message: "..." }) -- assign implementation from interaction blueprint
• close_agent({ target: "A11Y-001" }) -- cleanup after a11y audit

Error Handling