testgen-flow-gap-and-contradiction-analysis

<testgen_flow_gap_and_contradiction_analysis>

<description_and_purpose> Analyze the Issue Tracker ticket and Wiki documentation to identify contradictions, gaps, ambiguities, and inconsistencies that need clarification before requirements generation. </description_and_purpose>

<workflow_context>

• Phase 2 of 7 in testgen-flow
• Input: raw-data.md from Phase 1
• Output: analysis.md with categorized contradictions, gaps, ambiguities, risk assessment
• Required skills: qa-knowledge (gap_analysis mode)
• Prerequisite: Phase 0, Phase 1 complete </workflow_context>

<phase_steps>

1. Load raw data
2. Run gap and contradiction analysis
3. Create analysis document
4. Update state file </phase_steps>

<load_raw_data step="2.1">

1. Read plans/testgen-{TICKET-KEY}/raw-data.md completely
2. Extract key sections: ticket description and acceptance criteria, labels, components, priority, each Wiki page content, comments from both sources
3. Failure paths:
   • raw-data.md missing: stop Phase 2, record Phase 2 blocked: raw-data.md missing in testgen-state.md, and ask user to rerun Phase 1.
   • raw-data.md exists but key sections empty (no ticket description / no Wiki content): record the empty sections as gaps for Phase 3 to surface, and proceed — do not silently fabricate content.
   • raw-data.md corrupt / unparseable: stop Phase 2, record the parse error, and ask user to inspect the file. </load_raw_data>

<run_analysis step="2.2" subagent="architect" role="Requirements gap analyst">

1. USE SKILL qa-knowledge (gap_analysis mode, general multi-source variant). The mode is analysis-only and EMITS categorized findings into this phase's <analysis_document_contract> artifact; it never invents the artifact shape or path.
2. Sources to analyze: Issue Tracker ticket data + Wiki page data from raw-data.md.
3. Identify contradictions, gaps, ambiguities per the mode's detection catalogs — contradiction (value-mismatch / logic-conflict / requirement-conflict), gap (functional / non-functional / data / business-logic / dependency), and ambiguity (vague-term) probes. This phase does NOT restate the taxonomies; it invokes them through the mode and OWNS the output document below.
4. Cross-reference ticket vs Wiki for information present only in one source (single-source case → skip-with-note). </run_analysis>

<create_analysis_document step="2.3">

Create plans/testgen-{TICKET-KEY}/analysis.md. The qa-knowledge gap_analysis mode EMITS its categorized findings into the phase-owned document contract below — this phase OWNS the full skeleton, section list, and risk-assessment artifact shape; the mode supplies the finding entries.

Precondition (mode produced findings): step 2.2 invoked the gap_analysis mode and produced categorized findings (or an explicit zero-issues result). If the mode could not run, apply <failure_handling> "gap_analysis produced no findings" — do NOT fabricate a partial analysis.

<analysis_document_contract>

The document has these sections in order; empty finding sections carry No issues found (never silently omitted). Per-entry shapes: C[N] (Type / Source 1 / Source 2 / Impact / Needs Clarification), G[N] (Type / Context / Missing Information / Impact / Suggested Question), A[N] (Source / Vague Statement / Possible Interpretations ≥2 / Clarification Needed). Risk tiers are exactly three (High / Medium / Low) — no fourth tier. Phase 3 priority mapping (so the downstream P0/P1 signal is unambiguous): a High finding whose Impact is "blocks implementation" → Phase 3 P0 (MUST answer); Medium (impacts quality) → P1; Low (minor clarification) → P2/P3. Phase 3 classifies a question's priority from the finding's tier + its stated Impact, not from a tier name alone.

# Analysis - [TICKET-KEY]

**Analyzed**: [DateTime]
**Sources**: [Issue Tracker ticket + Wiki pages analyzed]

---

## Executive Summary
- **Total Issues Found**: [Count]
- **Contradictions**: [Count]  · **Gaps**: [Count]  · **Ambiguities**: [Count]
- **Severity**: [High / Medium / Low]
- **Recommendation**: [Can proceed with clarifications / Needs major rework]

## 1. Contradictions
[None found OR C[N] entries]

## 2. Gaps
[None found OR G[N] entries]

## 3. Ambiguities
[None found OR A[N] entries]

## 4. Cross-Reference Analysis
[Findings OR `Skipped — only one source available (<name>); no cross-reference possible.`]

## 5. Positive Findings
[Well-documented areas / strengths]

## 6. Risk Assessment
**High Risk** (blocks implementation): [Issue ID — why blocking]
**Medium Risk** (impacts quality): [Issue ID — impact]
**Low Risk** (minor clarification): [Issue ID — minor impact]

## 7. Next Steps
1. Generate clarification questions (Phase 3)
2. Total questions expected: [Estimate based on issues found]
3. Recommended: Review with [Stakeholder role] before proceeding

## Analysis Metadata
- **Ticket Fields Analyzed**: [List key fields]
- **Wiki Pages Analyzed**: [Count and titles]
- **Analysis Duration**: [Time spent]
- **Manual Review**: [Areas requiring human judgment]