feedback-recall by yohayetsion/product-org-os

Search and synthesize past feedback to inform current work.

Vision to Value Phase

Cross-phase - This skill surfaces customer voice before work in any phase.

Prerequisites: Feedback captured in the registry Outputs used by: All phases (ensures customer-informed decisions)

Purpose

Before making decisions, developing features, or analyzing markets, recall what customers and stakeholders have already told us. This skill surfaces relevant past feedback with its analysis and patterns.

When to Use

Invoke /feedback-recall [query] when:

Starting work on a feature (what have customers said about this area?)
Making a decision (what feedback supports or challenges this direction?)
Preparing for customer conversations (what have they told us before?)
Analyzing a market segment (what patterns exist in their feedback?)
Validating assumptions (what evidence do we have?)
Investigating a problem (what related complaints exist?)

Process

1. Parse the Query

Accept various query types, with optional filters:

Topic: /feedback-recall onboarding → feedback about onboarding
Feature: /feedback-recall API integration → feedback about API
Segment: /feedback-recall enterprise → feedback from enterprise customers
Source: /feedback-recall Acme Corp → feedback from specific customer
Sentiment: /feedback-recall negative pricing → negative pricing feedback
Theme: /feedback-recall TH-005 → feedback linked to a specific theme
Product: /feedback-recall onboarding product:AXIA → filtered to AXIA product
Demo: /feedback-recall onboarding --include-demo → include demo data

Filters:

product:[name] - Filter to specific product
--include-demo - Include demo data (marked with [DEMO])
--demo-only - Show only demo data (for testing/learning)

1b. Check for Production Data (Demo Filtering)

Before searching, determine if production data exists:

Check main context folders (NOT context/demo/):
- context/feedback/index.md - Any non-demo entries?
- Any files in context/feedback/[YYYY]/ that aren't demo?

Apply demo filtering rule:

Production Data?	Flag	Behavior
No	(any)	Include demo with `[DEMO]` markers
Yes	(none)	Exclude demo data, show excluded count
Yes	`--include-demo`	Include demo with `[DEMO]` markers
(any)	`--demo-only`	Only demo data

Demo data is identified by:
- Path contains context/demo/
- ID contains "DEMO" (e.g., FB-DEMO-001)

2. Search Feedback Registry

Read context/feedback/index.md and search for:

Topic tag matches
Source name matches
Product/feature matches
Segment matches
Sentiment matches

For strong matches, read the full feedback entry from context/feedback/[YYYY]/.

3. Check Themes

Read context/feedback/themes.md for:

Established themes related to the query
Theme status and trend information
Aggregated insights across multiple feedback entries

4. Synthesize Results

## Feedback Recall: [Query]

*Found [N] feedback entries related to "[query]"*

### Summary
[2-3 sentence synthesis of what feedback tells us about this topic]

### Sentiment Overview
| Sentiment | Count | Trend |
|-----------|-------|-------|
| Positive | [N] | [↑/↓/→] |
| Negative | [N] | [↑/↓/→] |
| Neutral | [N] | [↑/↓/→] |

### Key Themes

#### [TH-NNN]: [Theme Name]
- **Status**: [Status]
- **Frequency**: [N] mentions
- **Trend**: [Improving/Stable/Declining]
- **Summary**: [Theme summary]

### Representative Feedback

#### FB-[YYYY]-[NNN]: [Summary]
- **Source**: [Source] ([Segment])
- **Date**: [Date]
- **Sentiment**: [Sentiment]
- **Key Quote**: "[Quote]"
- **Insight**: [Key insight]

[Repeat for top 3-5 most relevant]

### All Related Feedback

| ID | Date | Source | Sentiment | Summary |
|----|------|--------|-----------|---------|
| [ID] | [Date] | [Source] | [Sent] | [Summary] |

### Patterns Observed
- [Pattern 1]
- [Pattern 2]
- [Pattern 3]

### Connections to Context

**Related Decisions**:
- [DR-IDs mentioned in feedback entries]

**Related Bets**:
- [SB-IDs mentioned in feedback entries]

**Assumption Evidence**:
- [A-ID]: [Supported/Challenged] by [N] feedback entries

### Recommendations

Based on this feedback:
1. [Recommendation 1]
2. [Recommendation 2]
3. [Recommendation 3]

### Gaps

Areas where we lack feedback:
- [Gap 1 - consider gathering more data]
- [Gap 2]

5. Highlight Actionable Insights

Call out:

Strong patterns that should inform decisions
Contradictions between feedback entries
Feedback that challenges current assumptions
Urgent issues requiring immediate attention
Opportunities identified across multiple sources

Instructions

Accept query from user (required)
Parse optional product:[name] filter from query
Read context/feedback/index.md
Read context/feedback/themes.md
Search for matches across all dimensions
If product filter specified, filter results to that product only
For top matches, read full feedback entries
Synthesize findings into actionable summary
Highlight patterns and themes
Note connections to decisions, bets, assumptions
Identify gaps where more feedback is needed
Provide recommendations based on feedback

Query Examples

/feedback-recall pricing
→ All feedback mentioning pricing, value, cost, ROI

/feedback-recall enterprise onboarding
→ Feedback from enterprise segment about onboarding

/feedback-recall negative
→ All negative sentiment feedback

/feedback-recall Acme Corp
→ All feedback from Acme Corp

/feedback-recall API
→ Feedback about API functionality, integrations

/feedback-recall Q4 2025
→ Feedback received in Q4 2025

/feedback-recall feature requests
→ All feature request type feedback

/feedback-recall pricing product:AXIA
→ Pricing feedback for AXIA product only

/feedback-recall product:SKYMOD
→ All feedback for SKYMOD product

If No Feedback Found

## Feedback Recall: [Query]

No feedback found matching "[query]".

This could indicate:
1. No feedback has been captured in this area yet
2. Try different keywords: [suggest alternatives]
3. This may be a gap in our customer intelligence

**Recommendation**: Consider gathering feedback in this area through:
- Customer interviews
- Sales team outreach
- Support ticket analysis
- Survey or research study

Graph-Enhanced Recall (v3)

When querying feedback:

Search feedback indexes: Use sourceIndex, sentimentIndex, topicIndex from context/index.json
Follow cross-references: Show decisions and bets that feedback links to
Show theme connections: If feedback is part of a theme, show the theme and other related feedback
Filter by product: In multi-product orgs, use productIndex for scoped queries

Integration with Other Skills

After /feedback-recall, consider /context-recall for related decisions
Feedback insights should inform /decision-record analysis
Feedback patterns can validate/invalidate /strategic-bet assumptions
Use findings to enhance /prd customer problem sections