/feedback                    # Same as status
/feedback status             # Show current state
/feedback pause              # Pause learning
/feedback resume             # Resume learning
/feedback reset              # Clear learned patterns
/feedback export             # Export feedback data
/feedback settings           # Show/edit settings
/feedback opt-in             # Enable anonymous sharing
/feedback opt-out            # Disable anonymous sharing
/feedback privacy            # View privacy policy
/feedback export-analytics   # Export anonymous analytics for review

Feedback System Status
-----------------------------
Learning: Enabled
Anonymous sharing: Disabled
Data retention: 90 days

Learned Patterns:
- Auto-approves: npm install, npm test, git push (3 commands)
- Code style: async/await preferred, TypeScript strict mode

Agent Performance:
- backend-architect: 94% success (28 spawns) [improving]
- test-generator: 72% success (18 spawns) [declining]

Context Savings: ~8k tokens/session (estimated)

Storage: .claude/feedback/ (45 KB)

Feedback learning paused

Your existing patterns are preserved.
Resume with: /feedback resume

Feedback learning resumed

The system will continue learning from your usage.

WARNING: This will clear all learned patterns:

- 5 auto-approve permission rules
- 3 code style preferences
- Agent performance history

Your preferences (enabled, sharing, retention) will be kept.

To confirm, respond with exactly: RESET
To cancel, respond with anything else.

Feedback data reset

- Cleared 5 permission patterns
- Cleared 3 style preferences
- Cleared agent metrics

Learning will start fresh from now.

Exported feedback data to:
   .claude/feedback/export-2026-01-14.json

Contains:
- 5 learned permission patterns
- 3 code style preferences
- 8 skill usage metrics
- 4 agent performance records

File size: 12 KB

Feedback Settings
-----------------------------
enabled:              true   (master switch)
learnFromEdits:       true   (learn from code edits)
learnFromApprovals:   true   (learn from permissions)
learnFromAgentOutcomes: true (track agent success)
shareAnonymized:      false  (share anonymous stats)
retentionDays:        90     (data retention period)

To change a setting, use:
  /feedback settings <key> <value>

Example:
  /feedback settings retentionDays 30

Anonymous analytics sharing enabled.

What we share (anonymized):
  - Skill usage counts and success rates
  - Agent performance metrics
  - Hook trigger counts

What we NEVER share:
  - Your code or file contents
  - Project names or paths
  - Personal information
  - mem0 memory data

Disable anytime: /feedback opt-out

Anonymous analytics sharing disabled.

Your feedback data stays completely local.
No usage data is shared.

Re-enable anytime: /feedback opt-in

═══════════════════════════════════════════════════════════════════
              ORCHESTKIT ANONYMOUS ANALYTICS PRIVACY POLICY
═══════════════════════════════════════════════════════════════════

WHAT WE COLLECT (only with your consent)
────────────────────────────────────────────────────────────────────

  ✓ Skill usage counts        - e.g., "api-design used 45 times"
  ✓ Skill success rates       - e.g., "92% success rate"
  ✓ Agent spawn counts        - e.g., "backend-architect spawned 8 times"
  ✓ Agent success rates       - e.g., "88% tasks completed successfully"
  ✓ Hook trigger counts       - e.g., "git-branch-protection triggered 120 times"
  ✓ Hook block counts         - e.g., "blocked 5 potentially unsafe commands"
  ✓ Plugin version            - e.g., "4.12.0"
  ✓ Report date               - e.g., "2026-01-14" (date only, no time)


WHAT WE NEVER COLLECT
────────────────────────────────────────────────────────────────────

  ✗ Your code or file contents
  ✗ Project names, paths, or directory structure
  ✗ User names, emails, or any personal information
  ✗ IP addresses (stripped at network layer)
  ✗ mem0 memory data or conversation history
  ✗ Architecture decisions or design documents
  ✗ API keys, tokens, or credentials
  ✗ Git history or commit messages
  ✗ Any data that could identify you or your projects


YOUR RIGHTS
────────────────────────────────────────────────────────────────────

  • Opt-out anytime:     /feedback opt-out
  • View your data:      /feedback export-analytics
  • Check status:        /feedback status
  • View this policy:    /feedback privacy

/feedback export-analytics [path]

Analytics exported to: .claude/feedback/analytics-exports/analytics-export-20260114-120000.json

Contents preview:
-----------------
Date: 2026-01-14
Plugin Version: 4.12.0

Summary:
  Skills used: 8
  Skill invocations: 45
  Agents used: 3
  Agent spawns: 12
  Hooks configured: 5

Please review the exported file before sharing.

{
  "timestamp": "2026-01-14",
  "plugin_version": "4.12.0",
  "skill_usage": {
    "api-design-framework": { "uses": 12, "success_rate": 0.92 }
  },
  "agent_performance": {
    "backend-system-architect": { "spawns": 8, "success_rate": 0.88 }
  },
  "hook_metrics": {
    "git-branch-protection": { "triggered": 45, "blocked": 3 }
  },
  "summary": {
    "unique_skills_used": 8,
    "unique_agents_used": 3,
    "hooks_configured": 5,
    "total_skill_invocations": 45,
    "total_agent_spawns": 12
  },
  "metadata": {
    "exported_at": "2026-01-14T12:00:00Z",
    "format_version": "1.0",
    "note": "Review before sharing"
  }
}

.claude/feedback/
├── preferences.json           # User preferences including shareAnonymized
├── consent-log.json          # GDPR consent audit trail (opt-in/opt-out history)
├── metrics.json              # Source data for analytics
└── analytics-exports/        # Exported analytics for review
    └── analytics-export-*.json