/synthesize-discovery

name: synthesize-discovery
description: Synthesize discovery findings into a consolidated analysis document with proposed tickets
argument-hint: <source-url> [source-url...] [--target=<epic-key>]

Synthesize Discovery Findings

**Arguments:** $ARGUMENTS

---

Argument Parsing

Parse the arguments to extract:

• `{Source_URLs}` - One or more Confluence document URLs (space-separated)
• `{Target_Epic}` - Optional target implementation epic (via `--target=CC-XX`)

**Examples:**

• `/synthesize-discovery https://confluence/doc1` → Single source, auto-detect target epics
• `/synthesize-discovery https://confluence/doc1 https://confluence/doc2` → Multiple sources
• `/synthesize-discovery https://confluence/doc1 --target=CC-62` → Single source, specific target epic
• `/synthesize-discovery doc1-url doc2-url doc3-url --target=CC-62` → Multiple sources, specific target

**Supported Source Types:**

• Discovery Documents (from `/create-epic-discovery`)
• Research findings documents
• Interview summaries
• Technical spike reports
• Competitive analysis documents
• Any Confluence page with structured findings

---

Workflow Chain

/create-epic-discovery <epic-key>  → Discovery Document
         ↓
[Manual research, interviews, experiments]
         ↓
/synthesize-discovery <doc-urls...>  → Synthesis Document (YOU ARE HERE)
         ↓
/approve-synthesis <synthesis-url>  → Creates Jira Tickets
         ↓
/create-implementation-plan <overview-doc>  → Implementation planning continues

---

Phase 0: Source Retrieval

1. **Fetch all source documents** from the provided URLs

2. **Extract from each source:**

• Document type (discovery doc, research findings, spike report, etc.)
• Key findings and insights
• Validated/invalidated hypotheses
• Answered research questions
• Remaining unknowns
• Recommendations

3. **Identify target implementation epics:**

• If `--target` specified, use that epic
• Otherwise, extract from "Target Implementation Epics" sections
• If none found, ask user to specify

4. **FAILURE CONDITION - Missing Information:**

If sources cannot be fetched or no target epics identified:

**STOP and prompt the user:**

   I was unable to retrieve required information.

   Sources Retrieved: [count]/[total]
   Failed Sources: [list URLs that failed]
   Target Epics Found: [list or "None"]

   Please provide:
   1. Confirm source URLs are accessible
   2. Target implementation epic(s): [e.g., CC-62, CC-60]

**DO NOT PROCEED** until confirmed.

5. **MANDATORY CHECKPOINT - Sources Confirmation:**

   Please confirm before I proceed:

   Sources to Synthesize:
   1. [URL] - [Document Title] ([type])
   2. [URL] - [Document Title] ([type])
   ...

   Target Implementation Epic(s): [list]

   Is this correct? (Yes / No / Correct)

**DO NOT PROCEED** without explicit user confirmation.

---

Phase 1: Cross-Source Analysis

1. **Consolidate findings across all sources:**

• Merge hypothesis validation results
• Combine research question answers
• Identify consistent themes
• Note contradictions or conflicts

2. **Create findings inventory:**

| Finding ID | Source(s) | Category | Finding | Confidence | Actionable | |------------|-----------|----------|---------|------------|------------| | F1 | Doc1, Doc2 | User Need | Users want X | High | Yes | | F2 | Doc1 | Technical | API supports Y | Med | Yes | | F3 | Doc3 | Business | ROI is Z | Low | Needs validation |

3. **Identify patterns:**

• Recurring themes across sources
• Convergent conclusions
• Divergent opinions requiring resolution

4. **Catalog remaining unknowns:**

• Questions still unanswered
• New questions that emerged
• Areas needing further research

---

Phase 2: Recommendation Generation

1. **Generate feature recommendations:** For each actionable finding, determine:

• Should this become a ticket?
• What type of work is it? (Feature, Enhancement, Spike, Research)
• Which implementation epic does it belong to?
• What's the priority based on findings?
• Are there dependencies on other recommendations?

2. **Create recommendation matrix:**

| Rec ID | Based On | Title | Type | Target Epic | Priority | Dependencies | |--------|----------|-------|------|-------------|----------|--------------| | R1 | F1, F2 | Implement X feature | Story | CC-62 | High | None | | R2 | F3 | Validate ROI assumption | Spike | CC-60 | Med | R1 | | R3 | F1 | Design user flow for X | Task | CC-62 | High | None |

3. **Identify scope decisions needed:**

• Features that could go multiple directions
• Trade-offs requiring stakeholder input
• MVP vs future phase decisions

---

Phase 3: Synthesis Document Creation

Create a comprehensive synthesis document:

1. Synthesis Overview

• **Discovery Epic:** Link to original discovery epic
• **Sources Analyzed:** List all source documents with links
• **Synthesis Date:** When this synthesis was created
• **Target Implementation Epics:** Links to target epics

2. Executive Summary

• **Key Insight:** One-paragraph summary of most important discovery
• **Recommendation:** High-level direction based on findings
• **Confidence Level:** Overall confidence in recommendations
• **Major Decisions Needed:** List any blocking decisions

3. Consolidated Findings

Validated Hypotheses

| ID | Original Hypothesis | Validation Status | Evidence | Implications | |----|---------------------|-------------------|----------|--------------| | H1 | [statement] | Validated | [sources] | [what this means] | | H2 | [statement] | Partially Validated | [sources] | [caveats] | | H3 | [statement] | Invalidated | [sources] | [pivot needed] |

Research Questions Answered

| ID | Question | Answer | Confidence | Source(s) | |----|----------|--------|------------|-----------| | Q1 | [question] | [answer] | High/Med/Low | [docs] |

Key Insights

Narrative s