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

Phase Swarm Decision When

Ideation qcsd-ideation-swarm GO / CONDITIONAL / NO-GO PI/Sprint Planning

Refinement qcsd-refinement-swarm READY / CONDITIONAL / NOT-READY Sprint Refinement

Development qcsd-development-swarm SHIP / CONDITIONAL / HOLD During Sprint

Verification qcsd-cicd-swarm RELEASE / REMEDIATE / BLOCK Pre-Release / CI-CD

Production qcsd-production-swarm HEALTHY / DEGRADED / CRITICAL Post-Release

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

Rule Enforcement

E1 You MUST spawn ALL THREE core agents (qe-quality-criteria-recommender, qe-risk-assessor, qe-requirements-validator) in Step 2. No exceptions.

E2 You MUST put all parallel Task calls in a SINGLE message.

E3 You MUST STOP and WAIT after each batch. No proceeding early.

E4 You MUST spawn conditional agents if flags are TRUE. No skipping.

E5 You MUST apply GO/CONDITIONAL/NO-GO logic exactly as specified in Step 5.

E6 You MUST generate the full report structure. No abbreviated versions.

E7 Each agent MUST read its reference files before analysis.

E8 You MUST execute Step 7 learning persistence. No skipping.

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

• 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)

• Core Agents -- steps/02-core-agents.md -- Spawn qe-quality-criteria-recommender, qe-risk-assessor, qe-requirements-validator in parallel

• Batch 1 Results -- steps/03-batch1-results.md -- Wait for core agents, extract all metrics

• Conditional Agents -- steps/04-conditional-agents.md -- Spawn flagged conditional agents in parallel

• Decision Synthesis -- steps/05-decision-synthesis.md -- Apply GO/CONDITIONAL/NO-GO logic

• Report Generation -- steps/06-report-generation.md -- Generate executive summary and full report

• Learning Persistence -- steps/07-learning-persistence.md -- Store findings to memory, save persistence record, include video follow-up recommendation

• Final Output -- steps/08-final-output.md -- Display completion summary with all scores and follow-up recommendations

Execution Instructions

• 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" }) )

• Execute the step's instructions completely

• Verify all success criteria are met before proceeding

• Pass the step's output as context to the next step

• 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

Agent Type Domain Batch

qe-quality-criteria-recommender Core (always) requirements-validation 1

qe-risk-assessor Core (always) coverage-analysis 1

qe-requirements-validator Core (always) requirements-validation 1

qe-accessibility-auditor Conditional (HAS_UI) visual-accessibility 2

qe-security-auditor Conditional (HAS_SECURITY) security-compliance 2

qe-qx-partner Conditional (HAS_UX) cross-domain 2

qe-middleware-validator Conditional (HAS_MIDDLEWARE) enterprise-integration 2

qe-sap-rfc-tester Conditional (HAS_SAP_INTEGRATION) enterprise-integration 2

qe-sod-analyzer Conditional (HAS_AUTHORIZATION) enterprise-integration 2

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

Metric GO CONDITIONAL NO-GO

Testability Score

= 70% 40 - 69% < 40%

Risk Score LOW/MEDIUM HIGH (mitigatable) CRITICAL

Requirements Completeness

= 80% 50 - 79% < 50%

Quality Criteria Coverage

= 7/10 categories 4-6/10 categories < 4/10 categories

Report Filename Mapping

Agent Report Filename Step

qe-quality-criteria-recommender 02-quality-criteria.md

2

qe-risk-assessor 03-risk-assessment.md

2

qe-requirements-validator 04-requirements-validation.md

2

qe-accessibility-auditor 05-accessibility-audit.md

4

qe-security-auditor 06-security-audit.md

4

qe-qx-partner 07-ux-assessment.md

4

qe-middleware-validator 08-middleware-assessment.md

4

qe-sap-rfc-tester 09-sap-assessment.md

4

qe-sod-analyzer 10-sod-compliance.md

4

Learning Persistence 11-learning-persistence.json

7

Synthesis 01-executive-summary.md

6

Execution Model Options

Model When to Use Agent Spawn

Task Tool (PRIMARY) Claude Code sessions Task({ subagent_type, run_in_background: true })

MCP Tools MCP server available fleet_init({}) / task_submit({})

CLI Terminal/scripts swarm init / agent spawn

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.