Spec Driven Feedback

Spec-Driven Feedback

Capture retrospective feedback from development workflows to improve processes, identify patterns, and enable continuous improvement — with structural anti-skip enforcement.

If ambiguous or conflicts detected: HALT and use AskUserQuestion

Execution Model

This skill expands inline. After invocation, execute Phase 00 Initialization immediately. Do not wait passively, ask permission, or offer execution options.

Self-Check (if ANY box is true = VIOLATION):

• Stopping to ask about token budget
• Stopping to offer execution options
• Waiting passively for user to say "go"
• Asking "should I execute this?"
• Skipping a phase because it "seems simple"
• Combining multiple phases into one
• Summarizing instead of loading a reference file
• Skipping verification because "I already wrote the file"

IF any box checked: EXECUTION MODEL VIOLATION. Go directly to Phase 00 Initialization now.

Anti-Skip Enforcement Contract

This skill enforces 4 independent anti-skip layers. ALL FOUR must fail for a step to be skipped:

1. Per-phase reference loading - Each phase loads its reference files fresh via Read(). NOT consolidated. Prevents "already covered" rationalization.
2. Binary CLI gates - devforgeai-validate phase-check/phase-complete at phase boundaries. Cannot be forged by LLM.
3. Checkpoint-based state tracking - Phase completion verified by checking checkpoint JSON data keys and current_phase field.
4. Artifact verification - Checkpoint JSON existence checked via Glob(), feedback files verified on disk, index entries verified via Grep().

Execute-Verify-Record Pattern: Every mandatory step in every phase file has three parts:

• EXECUTE: The exact action to perform (Read, Write, AskUserQuestion, Task, Grep, Glob)
• VERIFY: How to confirm the action happened (file exists, content contains expected text, data key populated)
• RECORD: Update checkpoint JSON with captured data; call devforgeai-validate phase-record

Token Optimization Bias is PROHIBITED. Do not skip, compress, or shortcut any step. Every phase step exists because a previous failure proved it necessary.

Feedback Types

All references above are located at references/ (relative to this skill directory).

Parameter Extraction

Extract from conversation context markers set by invoking command:

Command Integration

These commands delegate to this skill. When invoked via a command, context markers are already set.

Hook Integration

This skill is auto-invoked by the event-driven hook system (STORY-018):

Hook configuration: devforgeai/config/hooks.yaml Hook system reference: HOOK-SYSTEM.md (relative to this skill directory)

Phase 00: Initialization [INLINE - Bootstraps State]

This phase runs inline because it creates the state that all other phases depend on.

Step 0.1: Parse Arguments

Extract from conversation context markers:
  $FEEDBACK_MODE     — from "**Feedback Mode:**" marker or default "conversation"
  $FEEDBACK_CONTEXT  — from "**Feedback Context:**" marker or extract from conversation
  $FEEDBACK_SOURCE   — from "**Feedback Source:**" marker or default "manual"
  $SEARCH_QUERY      — from "**Search Query:**" marker (search operations)
  $SEVERITY          — from "**Severity:**" marker (search/export)
  $STATUS            — from "**Status:**" marker (search)
  $PRIORITY_FILTER   — from "**Priority Filter:**" marker (triage)
  $SELECTED_ITEMS    — from "**Selected Items:**" marker (triage)
  $FORMAT            — from "**Format:**" marker (export)
  $DATE_RANGE        — from "**Date Range:**" marker (export)
  $STORY_IDS         — from "**Story IDs:**" marker (export)
  $SUBCOMMAND        — from "**Subcommand:**" marker (config)
  $SANITIZE          — from "**Sanitize:**" marker (export ZIP)
  $OUTPUT_PATH       — from "**Output:**" marker (export ZIP)
  $ARCHIVE_PATH      — from "**Archive:**" marker (import ZIP)

Step 0.2: Resume Detection

IF conversation mentions "--resume" or resume context:
  Glob(pattern="devforgeai/feedback/checkpoints/FB-*.checkpoint.json")
  IF matching checkpoint found:
    Read the checkpoint file
    AskUserQuestion:
      Question: "Found existing feedback session. Resume or start fresh?"
      Header: "Resume"
      Options:
        - label: "Resume session"
          description: "Continue from last checkpoint"
        - label: "Start fresh"
          description: "Begin new feedback session"
    IF "Resume": Restore state, GOTO Phase Orchestration Loop at CURRENT_PHASE
ELSE:
  Continue to Step 0.3

Step 0.3: Generate Session ID

# Scan for highest existing FB-YYYY-MM-DD-### in register
register_exists = Glob(pattern="devforgeai/feedback/feedback-register.md")

IF register_exists:
  Read(file_path="devforgeai/feedback/feedback-register.md")
  Extract highest ### for today's date
  SESSION_ID = "FB-{YYYY-MM-DD}-{###+1}" (zero-padded to 3 digits)
ELSE:
  SESSION_ID = "FB-{YYYY-MM-DD}-001"

Step 0.4: CLI Initialization

devforgeai-validate phase-init ${SESSION_ID} --workflow=feedback --project-root=.

Step 0.5: Create Initial Checkpoint

checkpoint = {
  "checkpoint_version": "1.0",
  "session_id": SESSION_ID,
  "workflow": "feedback",
  "created_at": "current ISO 8601 timestamp",
  "updated_at": "current ISO 8601 timestamp",
  "status": "in_progress",
  "input": {
    "feedback_mode": $FEEDBACK_MODE,
    "feedback_source": $FEEDBACK_SOURCE,
    "feedback_context": $FEEDBACK_CONTEXT,
    "story_id": extracted story ID or null,
    "operation_type": extracted operation or null,
    "operation_status": extracted status or null,
    "search_query": $SEARCH_QUERY or null,
    "severity": $SEVERITY or null,
    "status_filter": $STATUS or null,
    "priority_filter": $PRIORITY_FILTER or null,
    "selected_items": $SELECTED_ITEMS or null,
    "format": $FORMAT or null,
    "date_range": $DATE_RANGE or null,
    "subcommand": $SUBCOMMAND or null
  },
  "progress": {
    "current_phase": 0,
    "phases_completed": [],
    "total_steps_completed": 0
  },
  "phases": {
    "01": { "status": "pending", "steps_completed": [] },
    "02": { "status": "pending", "steps_completed": [] },
    "03": { "status": "pending", "steps_completed": [] },
    "04": { "status": "pending", "steps_completed": [] },
    "05": { "status": "pending", "steps_completed": [] },
    "06": { "status": "pending", "steps_completed": [] }
  },
  "output": {
    "feedback_id": null,
    "feedback_file_path": null,
    "feedback_type": null,
    "stories_created": [],
    "error": null
  }
}

Write(file_path="devforgeai/feedback/checkpoints/${SESSION_ID}.checkpoint.json", content=checkpoint)

VERIFY: Glob(pattern="devforgeai/feedback/checkpoints/${SESSION_ID}.checkpoint.json") IF not found: HALT -- "Initial checkpoint was NOT created."

Step 0.6: Display Session Banner

Display:
"━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
  DevForgeAI Feedback Session
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Session: ${SESSION_ID}
Mode: ${FEEDBACK_MODE}
Source: ${FEEDBACK_SOURCE}
Context: ${FEEDBACK_CONTEXT || 'None provided'}
Story: ${story_id || 'N/A'}

Phases: 6 (Context Detection > Type Dispatch > Feedback Execution > Validation > Persistence > Completion)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━"

Set CURRENT_PHASE = 1.

Phase Orchestration Loop

FOR phase_num in range(CURRENT_PHASE, 7):  # Phases 01-06

    1. ENTRY GATE: devforgeai-validate phase-check ${SESSION_ID} --workflow=feedback --from={prev} --to={phase_id} --project-root=.
       IF exit != 0 AND exit != 127: HALT

    2. LOAD: Read(file_path=".claude/skills/spec-driven-feedback/phases/{phase_files[phase_num]}")
       Load the phase file FRESH. Do NOT rely on memory of previous reads.

    3. REFERENCE: Read the phase's reference files as specified in the phase Contract section.
       References are in references/ (self-contained within this skill).
       Load ALL listed references. Do not skip any.

    4. EXECUTE: Follow EVERY step in the phase file using EXECUTE-VERIFY-RECORD triplets.
       - Each step's EXECUTE tells you exactly what action to take
       - Each step's VERIFY tells you how to confirm the action happened
       - Each step's RECORD tells you how to update the checkpoint

    5. EXIT GATE: devforgeai-validate phase-complete ${SESSION_ID} --workflow=feedback --phase={phase_id} --checkpoint-passed --project-root=.
       IF exit != 0 AND exit != 127: HALT

    6. CHECKPOINT: Update checkpoint JSON with phase completion.
       Write updated checkpoint to disk.
       Verify write via Glob().

Phase Table

Required Subagents Per Phase

All other phases: No subagents required. Direct tool calls (Read, Write, Glob, Grep, AskUserQuestion).

AI Analysis Output Schema

{
  "story_id": "STORY-XXX",
  "timestamp": "ISO8601",
  "ai_analysis": {
    "what_worked_well": [{"observation": "...", "evidence": "...", "impact": "..."}],
    "areas_for_improvement": [{"issue": "...", "evidence": "...", "root_cause": "..."}],
    "recommendations": [{
      "title": "...",
      "description": "...",
      "affected_files": ["..."],
      "implementation_code": "...",
      "effort_estimate": "15 min|30 min|1 hour|2 hours|4 hours",
      "priority": "HIGH|MEDIUM|LOW",
      "feasible_in_claude_code": true
    }],
    "patterns_observed": ["..."],
    "anti_patterns_detected": ["..."],
    "constraint_analysis": "..."
  }
}

Constraint: All recommendations MUST be implementable within Claude Code Terminal. If a recommendation requires tools beyond Read, Write, Edit, Glob, Grep, Bash, TaskCreate, TaskUpdate, AskUserQuestion — flag it as feasible_in_claude_code: false.

State Persistence

• Checkpoint: devforgeai/feedback/checkpoints/${SESSION_ID}.checkpoint.json
• References: references/ (self-contained within this skill)
• Templates: templates/ (self-contained within this skill)
• Hook System: HOOK-SYSTEM.md (self-contained within this skill)
• Configuration: devforgeai/feedback/config.yaml

Workflow Completion Validation

IF phases_completed < 6: HALT "WORKFLOW INCOMPLETE - {completed_count}/6 phases"
IF feedback_file_path is null: HALT "Feedback file was not written"
IF feedback_id is null: HALT "Feedback ID was not generated"
IF checkpoint status != "complete": Update checkpoint status to "complete"

Configuration

File: devforgeai/feedback/config.yaml

Success Criteria

• All 6 phases executed (no skipping)
• Feedback file written to disk (verified via Glob)
• Index updated with new entry (verified via Grep)
• Register updated with new entry (verified via Grep)
• Checkpoint marked as complete
• Confirmation displayed to user with feedback ID
• Token usage < 40K (isolated context)

Related Documentation

• Original Skill: .claude/skills/spec-driven-feedback/SKILL.md (archived — absorbed into spec-driven-feedback per ADR-040)
• Hook System: HOOK-SYSTEM.md (self-contained within this skill)
• Hook Config: devforgeai/config/hooks.yaml
• Feedback Config: devforgeai/feedback/config.yaml
• Framework Analyst: .claude/agents/framework-analyst.md