📋 Work — Plan Execution Engaged
Track tasks, manage plans, and survive CC compacts with three operating modes.
Activation
When this skill activates, output:
📋 Work — Plan execution engaged.
Then determine which mode to use based on the trigger.
Context Guard
| Context |
Status |
| User says "copy plan", "append plan", "resume plan" |
ACTIVE — use matching mode |
| User says "what's next", "todo", "priorities" |
ACTIVE — quick query mode |
| User provides a task list or plan |
ACTIVE — copy mode |
| General discussion about planning concepts |
DORMANT — do not activate |
| User is executing a task (not managing the list) |
DORMANT — do not activate |
Step 0: Silent Context Compilation (MANDATORY)
Before executing ANY mode, silently gather current state. Do NOT present findings. Do NOT ask questions. Just internalize:
- Read the project's
STATE.md (if it exists) — current task, blockers, next steps
- Read the project's
CLAUDE.md (if it exists) — conventions, architecture decisions
- Check recent diary:
python C:/Projects/memstack/db/memstack-db.py get-sessions <project> --limit 3
- Check git state:
git log --oneline -5
git diff --stat
This is silent. Synthesize an internal understanding of where the project stands. Then proceed to the triggered mode with full context. The user already knows their project state — don't waste their time repeating it back.
Mode 1: Copy Plan
Trigger: "copy plan" or when a new plan is provided
- Parse the entire plan into individual numbered tasks
- For each task, save to SQLite:
python C:/Projects/memstack/db/memstack-db.py add-plan-task '{"project":"<name>","task_number":<n>,"description":"<task>","status":"pending"}'
- Confirm with task count
- Also write a markdown copy to
memory/projects/{project}-plan.md for human readability
Status values: pending, in_progress, completed, blocked
Mode 2: Append Plan (Update Tasks)
Trigger: "append plan" or when updating task statuses
- Read current plan from SQLite:
python C:/Projects/memstack/db/memstack-db.py get-plan <project>
- Update individual task statuses:
python C:/Projects/memstack/db/memstack-db.py update-task '{"project":"<name>","task_number":<n>,"status":"completed"}'
- Add new tasks if needed via
add-plan-task
- No size limits needed — SQLite handles scale
Mode 3: Resume Plan
Trigger: "resume plan" — use after CC compact or new session
- Load plan from SQLite:
python C:/Projects/memstack/db/memstack-db.py get-plan <project>
- Parse the JSON output for task statuses
- Output a summary:
Plan: {project} ({done}/{total} complete)
Completed: [list]
In Progress: [list]
Pending: [list]
Blocked: [list with reasons]
Recommended next: {first pending task}
- Continue from the first incomplete task
Quick Commands
- "what's next" — queries plan, returns the single next pending task
- "priorities" — shows top 3 pending items from the plan
- "todo" — shows all pending and in-progress items with status
Inputs
- Plan text (copy mode) or project name (append/resume)
- Database:
C:\Projects\memstack\db\memstack.db (via memstack-db.py)
- Fallback:
C:\Projects\memstack\memory\projects\ (legacy markdown)
Outputs
- Formatted task list with status indicators
- Updated plan file in memory
- Next recommended action
Example Usage
User: "resume plan for AdminStack"
📋 Work — Plan execution engaged.
Plan: AdminStack (5/9 complete)
[x] 1. Build CC Monitor page
[x] 2. Add setup guide
[x] 3. Fix API key validation
[x] 4. Add refresh feedback
[x] 5. Update guide with curl snippet
[ ] 6. Build cc-reporter.js Node script
[ ] 7. Add WebSocket real-time updates
[ ] 8. Session grouping by project
[!] 9. Mobile polish (blocked: waiting for design specs)
Recommended next: Task 6 — Build cc-reporter.js
Level History
- Lv.1 — Base: Single-mode TODO tracking. (Origin: MemStack v1.0, Feb 2026)
- Lv.2 — Enhanced: Three modes (copy/append/resume), 1K-line limit with auto-summarize, context guard, YAML frontmatter. (Origin: MemStack v2.0 MemoryCore merge, Feb 2026)
- Lv.3 — Advanced: SQLite-backed plans with per-task status tracking, no size limits, structured queries. (Origin: MemStack v2.1 Accomplish-inspired upgrade, Feb 2026)
- Lv.4 — Native: CC rules integration (
.claude/rules/work.md), always-on task planning awareness without skill file read. (Origin: MemStack v3.0-beta, Feb 2026)
- Lv.5 — Context-aware: Silent context compilation (Step 0) — reads STATE.md, CLAUDE.md, recent diary, and git state before any operation. Inspired by Intellegix silent pre-flight pattern. (Origin: MemStack v3.2, Feb 2026)
1---2name: work3description: Use when the user says 'plan', 'todo', 'copy plan', 'append plan', 'resume plan', 'priorities', or 'what's next'.4---567# 📋 Work — Plan Execution Engaged8*Track tasks, manage plans, and survive CC compacts with three operating modes.*910## Activation1112When this skill activates, output:1314`📋 Work — Plan execution engaged.`1516Then determine which mode to use based on the trigger.1718## Context Guard1920| Context | Status |21|---------|--------|22| **User says "copy plan", "append plan", "resume plan"** | ACTIVE — use matching mode |23| **User says "what's next", "todo", "priorities"** | ACTIVE — quick query mode |24| **User provides a task list or plan** | ACTIVE — copy mode |25| **General discussion about planning concepts** | DORMANT — do not activate |26| **User is executing a task (not managing the list)** | DORMANT — do not activate |2728## Step 0: Silent Context Compilation (MANDATORY)2930Before executing ANY mode, silently gather current state. Do NOT present findings. Do NOT ask questions. Just internalize:31321. Read the project's `STATE.md` (if it exists) — current task, blockers, next steps332. Read the project's `CLAUDE.md` (if it exists) — conventions, architecture decisions343. Check recent diary:35 ```bash36 python C:/Projects/memstack/db/memstack-db.py get-sessions <project> --limit 337 ```384. Check git state:39 ```bash40 git log --oneline -541 git diff --stat42 ```4344**This is silent.** Synthesize an internal understanding of where the project stands. Then proceed to the triggered mode with full context. The user already knows their project state — don't waste their time repeating it back.4546## Mode 1: Copy Plan4748**Trigger:** "copy plan" or when a new plan is provided49501. Parse the entire plan into individual numbered tasks512. For each task, save to SQLite:52 ```bash53 python C:/Projects/memstack/db/memstack-db.py add-plan-task '{"project":"<name>","task_number":<n>,"description":"<task>","status":"pending"}'54 ```553. Confirm with task count564. Also write a markdown copy to `memory/projects/{project}-plan.md` for human readability5758**Status values:** `pending`, `in_progress`, `completed`, `blocked`5960## Mode 2: Append Plan (Update Tasks)6162**Trigger:** "append plan" or when updating task statuses63641. Read current plan from SQLite:65 ```bash66 python C:/Projects/memstack/db/memstack-db.py get-plan <project>67 ```682. Update individual task statuses:69 ```bash70 python C:/Projects/memstack/db/memstack-db.py update-task '{"project":"<name>","task_number":<n>,"status":"completed"}'71 ```723. Add new tasks if needed via `add-plan-task`734. No size limits needed — SQLite handles scale7475## Mode 3: Resume Plan7677**Trigger:** "resume plan" — use after CC compact or new session78791. Load plan from SQLite:80 ```bash81 python C:/Projects/memstack/db/memstack-db.py get-plan <project>82 ```832. Parse the JSON output for task statuses843. Output a summary:85 ```86 Plan: {project} ({done}/{total} complete)8788 Completed: [list]89 In Progress: [list]90 Pending: [list]91 Blocked: [list with reasons]9293 Recommended next: {first pending task}94 ```954. Continue from the first incomplete task9697## Quick Commands9899- **"what's next"** — queries plan, returns the single next pending task100- **"priorities"** — shows top 3 pending items from the plan101- **"todo"** — shows all pending and in-progress items with status102103## Inputs104- Plan text (copy mode) or project name (append/resume)105- Database: `C:\Projects\memstack\db\memstack.db` (via memstack-db.py)106- Fallback: `C:\Projects\memstack\memory\projects\` (legacy markdown)107108## Outputs109- Formatted task list with status indicators110- Updated plan file in memory111- Next recommended action112113## Example Usage114115**User:** "resume plan for AdminStack"116117```118📋 Work — Plan execution engaged.119120Plan: AdminStack (5/9 complete)121122[x] 1. Build CC Monitor page123[x] 2. Add setup guide124[x] 3. Fix API key validation125[x] 4. Add refresh feedback126[x] 5. Update guide with curl snippet127[ ] 6. Build cc-reporter.js Node script128[ ] 7. Add WebSocket real-time updates129[ ] 8. Session grouping by project130[!] 9. Mobile polish (blocked: waiting for design specs)131132Recommended next: Task 6 — Build cc-reporter.js133```134135## Level History136137- **Lv.1** — Base: Single-mode TODO tracking. (Origin: MemStack v1.0, Feb 2026)138- **Lv.2** — Enhanced: Three modes (copy/append/resume), 1K-line limit with auto-summarize, context guard, YAML frontmatter. (Origin: MemStack v2.0 MemoryCore merge, Feb 2026)139- **Lv.3** — Advanced: SQLite-backed plans with per-task status tracking, no size limits, structured queries. (Origin: MemStack v2.1 Accomplish-inspired upgrade, Feb 2026)140- **Lv.4** — Native: CC rules integration (`.claude/rules/work.md`), always-on task planning awareness without skill file read. (Origin: MemStack v3.0-beta, Feb 2026)141- **Lv.5** — Context-aware: Silent context compilation (Step 0) — reads STATE.md, CLAUDE.md, recent diary, and git state before any operation. Inspired by Intellegix silent pre-flight pattern. (Origin: MemStack v3.2, Feb 2026)
vbrevik