/aisdlc-checkpoint-tasks - Save Task Progress

Checkpoint active tasks against current conversation context and update ACTIVE_TASKS.md.

<!-- Implements: REQ-TOOL-003 (Workflow Commands) --> <!-- Implements: REQ-TASK-001 (Work Breakdown) --> <!-- Implements: REQ-TOOL-002 (Developer Workspace) -->

Usage: /aisdlc-checkpoint-tasks

Instructions:

Phase 1: Analyze Current Context

1. Review conversation history to identify:
   • What work has been completed recently
   • What files were modified
   • What tests were run and their results
   • What commits were made
   • Any side tasks or tangential work done

Phase 2: Evaluate Active Tasks

1. Read .ai-workspace/tasks/active/ACTIVE_TASKS.md
2. For each active task, determine:
   • Completed: All acceptance criteria met, tests passing, documented
   • In Progress: Work has started, some acceptance criteria met
   • Blocked: Dependencies not met or blockers encountered
   • Not Started: No evidence of work in current context
   • Partially Relevant: Side work relates to this task

Phase 3: Update ACTIVE_TASKS.md

For each task:

For COMPLETED tasks:

1. Read .ai-workspace/templates/FINISHED_TASK_TEMPLATE.md
2. Create finished task document:
   • Path: .ai-workspace/tasks/finished/{YYYYMMDD_HHMM}_{task_slug}.md
   • Fill all sections based on conversation context
3. Remove completed task from ACTIVE_TASKS.md
4. Add to "Recently Completed" in Summary section
5. Update timestamp at top of file
6. Log: "✅ Task #{id} completed and archived"

For IN PROGRESS tasks:

1. Update status field to "In Progress"
2. Add progress notes (if any)
3. Update acceptance criteria checkboxes
4. Add any blockers discovered
5. Update timestamp at top of file
6. Log: "🔄 Task #{id} status updated to In Progress"

For BLOCKED tasks:

1. Update status to "Blocked"
2. Document blocker reason in task Notes
3. Update timestamp at top of file
4. Log: "🚫 Task #{id} marked as Blocked"

For NOT STARTED tasks:

1. No changes - leave as is
2. Log: "⏸️ Task #{id} not started (no changes)"

Phase 4: Summary Report

Provide a summary in this format:

╔══════════════════════════════════════════════════════════════╗
║              Task Checkpoint - {YYYY-MM-DD HH:MM}            ║
╚══════════════════════════════════════════════════════════════╝

📊 Evaluation Summary:
   ✅ Completed: {count} task(s)
   🔄 In Progress: {count} task(s)
   🚫 Blocked: {count} task(s)
   ⏸️  Not Started: {count} task(s)

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

✅ Completed Tasks:
   {list completed task IDs and titles with archive paths}

🔄 In Progress:
   {list in-progress task IDs, titles, and what was done}

🚫 Blocked:
   {list blocked task IDs, titles, and blocker reasons}

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

💾 Files Updated:
   - .ai-workspace/tasks/active/ACTIVE_TASKS.md
   {- .ai-workspace/tasks/finished/YYYYMMDD_HHMM_task_name.md (for each completed)}

💡 Next Steps:
   {suggestion based on remaining active tasks}

---

Notes:

• This command is particularly useful after side tasks or long work sessions
• It helps maintain accurate task state based on actual work done
• Use before /aisdlc-status to get accurate status
• The command uses conversation context, so ensure relevant work is in current context
• For ambiguous cases, ask the user for clarification before marking as completed

Example Use Cases:

• After working on a side task, checkpoint to see if any active tasks were incidentally completed
• At end of work session to update all task statuses
• Before context switch to ensure task state is accurate
• After a series of commits to update task progress