qcsd-ideation-swarm（qcsd 创意群）

QCSD Ideation Swarm v7.0

Shift-left quality engineering swarm for PI Planning and Sprint Planning.

Overview

The Ideation Swarm assesses stories and URLs before development begins, using HTSM v6.3 quality criteria, Risk Storming, and Testability analysis. It renders a GO / CONDITIONAL / NO-GO decision and is the first gate in the QCSD lifecycle.

QCSD Phase Positioning

Parameters

• URL: Website to analyze (required for URL mode)
• OUTPUT_FOLDER: Where to save reports (default: ${PROJECT_ROOT}/Agentic QCSD/{domain}/)

Unique Features

• URL Mode: Automated 5-tier browser cascade for live website analysis
• HAS_VIDEO Flag: Detects video content and recommends /a11y-ally follow-up
• HTML Output: Generates HTML reports when html_output: true
• Browser Cascade: Vibium -> agent-browser -> Playwright+Stealth -> WebFetch -> WebSearch-fallback

ENFORCEMENT RULES - READ FIRST

PROHIBITED BEHAVIORS:

• Summarizing instead of spawning agents
• Skipping agents "for brevity"
• Proceeding before background tasks complete
• Providing your own analysis instead of spawning specialists
• Omitting report sections or using placeholder text

Step Execution Protocol

This skill uses a micro-file step architecture. Each step is a self-contained file loaded one at a time to avoid "lost in the middle" context degradation.

Execute steps sequentially by reading each step file with the Read tool.

Steps

1. Flag Detection -- steps/01-flag-detection.md -- Fetch URL content (browser cascade), detect all 7 flags (HAS_UI, HAS_SECURITY, HAS_UX, HAS_VIDEO, HAS_MIDDLEWARE, HAS_SAP_INTEGRATION, HAS_AUTHORIZATION)
2. Core Agents -- steps/02-core-agents.md -- Spawn qe-quality-criteria-recommender, qe-risk-assessor, qe-requirements-validator in parallel
3. Batch 1 Results -- steps/03-batch1-results.md -- Wait for core agents, extract all metrics
4. Conditional Agents -- steps/04-conditional-agents.md -- Spawn flagged conditional agents in parallel
5. Decision Synthesis -- steps/05-decision-synthesis.md -- Apply GO/CONDITIONAL/NO-GO logic
6. Report Generation -- steps/06-report-generation.md -- Generate executive summary and full report
7. Learning Persistence -- steps/07-learning-persistence.md -- Store findings to memory, save persistence record, include video follow-up recommendation
8. Final Output -- steps/08-final-output.md -- Display completion summary with all scores and follow-up recommendations

Execution Instructions

1. Use the Read tool to load the current step file (e.g., Read({file_path: ".claude/skills/qcsd-ideation-swarm/steps/01-flag-detection.md"}))
2. Execute the step's instructions completely
3. Verify all success criteria are met before proceeding
4. Pass the step's output as context to the next step
5. If a step fails, halt and report the failure point -- do not skip ahead

Resume Support

To resume from a specific step: specify --from-step N and the orchestrator will skip to step N. Ensure you have the required prerequisite data from prior steps.

Agent Inventory

Total: 9 agents (3 core + 6 conditional)

Note: The Ideation swarm has 8 steps (not 9) because it has no separate final analysis agent step. The HAS_VIDEO flag triggers a follow-up recommendation for /a11y-ally but does not spawn an additional agent.

Quality Gate Thresholds

Report Filename Mapping

Execution Model Options

Key Principle

Quality is built in from the start, not tested in at the end. This swarm ensures stories are viable, testable, and risk-assessed before any code is written.