HITL: Approval queue, code diff view, graduation, confidence scoring, and notifications

[Copilot]

Formalizes the human-in-the-loop workflow that connects the Fix Review panel, pattern library, and self-healing mode into a cohesive system with persistence, graduation logic, and autonomous promotion.

Backend (serve.py)

Persistent Queue

• fix-queue.json / fix-approval-history.json survive restarts; all mutations guarded by _fixes_lock
• Fix records now carry: issue_class, confidence, created_at, outcome, changed_files, diffs, auto_applied

New endpoints

Method	Path	Purpose
POST	/api/fix/resolve	Approve / Partial / Reject; updates per-class history, triggers regression check
GET	/api/fix/diff	Parses session JSONL for Edit/Write tool calls → unified diff per file
POST	/api/fix/revert	git checkout -- <changed_files> for rejected fixes
GET	/api/fix/graduation-stats	Per-class approval rates and graduation eligibility
POST	/api/fix/graduate	Promotes class to autonomous (auto_apply: true)
POST	/api/fix/webhook	Configures Slack/webhook URL + sound toggle

Graduation & confidence

• compute_confidence() scales 0.3–1.0 from class approval history; novel classes default to 0.3
• check_graduation() requires ≥95% approval over ≥10 fixes
• Graduated classes with confidence ≥ 0.9 auto-apply and skip the HITL queue entirely
• Regression detection: a rejected fix from a graduated class revokes auto_apply

Frontend (js/fix-review.js)

• Issue-class badge + confidence badge (green ≥70%, orange <70%) on every fix card
• Expandable syntax-highlighted unified diff viewer (+ green / - red / @@ blue)
• Graduation bar appears when a class hits threshold; "Graduate → Autonomous" button calls POST /api/fix/graduate
• Collapsible graduation stats table with per-class outcomes and inline Graduate buttons
• Batch Approve All / Reject All when multiple pending fixes share a class
• Revert button activates after rejection, calls POST /api/fix/revert
• resolve() now persists via POST /api/fix/resolve (previously only called window.__claude.resolve)
• Desktop Notification API fires when a fix transitions to ready
• _escHtml() helper applied throughout to prevent XSS from untrusted fix metadata

CSS (style.css)

Adds 222 lines: confidence badges, unified-diff syntax highlighting, graduation bar/stats table, batch action bar, revert button.

Original prompt

---

This section details on the original issue you should resolve

<issue_title>HITL: Human-in-the-loop approval workflow for autonomous fixes</issue_title>
<issue_description>## Description
Formalize the human-in-the-loop approval workflow that connects the Fix Review panel (#10), pattern library (#8), and self-healing mode (#7) into a cohesive HITL system.

Current State (Delivered)

• Fix Review panel: Before/after screenshots, Approve/Partial/Reject buttons
• Pattern library: Stores successful fixes, matches on new issues
• Telemetry: resolve() closes the feedback loop, metrics() shows dashboard

HITL Requirements

1. Approval Queue

• Persistent queue of pending fixes (survives page refresh)
• Queue items show: description, issue class, before/after screenshots, code diff
• Batch approve/reject for same-class fixes
• Queue badge count in the visualizer header bar

2. Code Diff View

• Show the actual file edits made by Claude Code (parse session JSONL for Edit/Write tool calls)
• Syntax-highlighted diff alongside the visual before/after
• One-click revert if rejected (git checkout the changed files)

3. Approval Graduation (ADR-0007)

• Track approval rate per issue class
• When class hits 95%+ approval over 10+ fixes, propose auto-apply
• "Graduate" button promotes class from HITL → autonomous
• Graduated classes still log telemetry but skip the review queue
• Regression detection: if a graduated fix fails, revoke auto-apply

4. Confidence Scoring

• Display confidence score based on pattern match quality
• High-confidence fixes (known class, strong pattern match) show green badge
• Low-confidence fixes (novel class, no prior patterns) show orange badge
• Confidence threshold configurable (default: auto-apply above 0.9)

5. Notification Integration

• Desktop notification when fix is ready for review
• Optional sound (via TTS feedback skill)
• Slack/webhook notification for remote review

Architecture

__claude.fix() → Before screenshot → Claude Code session → After screenshot
                                                                    ↓
                                              Fix Review Panel (HITL)
                                              ├── Before/After images
                                              ├── Code diff
                                              ├── Confidence score
                                              └── [Approve] [Partial] [Reject]
                                                      ↓
                                              Pattern stored (success)
                                              → Metrics updated
                                              → Graduation check

Depends On

• ADR-0006: DevTools Bridge integration #6 (ADR-0006: DevTools Bridge) — Done
• Fix pattern library: store and reuse successful fix patterns #8 (Fix pattern library) — Done
• Before/after visual diff for fix approval workflow #10 (Before/after visual diff) — Done
• Self-healing UI: proactive issue detection and auto-fix queue #7 (Self-healing UI) — Open (provides the fix source)
• Context pruning: reduce context size using telemetry signal-to-noise data #9 (Context pruning) — Open (optimizes what gets sent)

References

• PKM ADR-0007: Always-On Agent Fleet (operational autonomy)
• PKM ADR-0013 Benchmarks: Self-Improvement Loop 3 (Approval Lane Graduation)
• PKM ADR-0013 Benchmarks: North Star principles</issue_description>

Comments on the Issue (you are @copilot in this section)

• Fixes HITL: Human-in-the-loop approval workflow for autonomous fixes #12

---

📱 Kick off Copilot coding agent tasks wherever you are with GitHub Mobile, available on iOS and Android.