mp-web3/claude-starter-kit/tasks
Claude Code and agent skill by mp-web3/claude-starter-kit
- community
Checking that the SKILL.md file is still in the repository.
Manage the task backlog — add, review, complete, and measure tasks against the 12 Favorite Problems. Run /tasks to see current status.
Install
npx vskill@latest install mp-web3/claude-starter-kit/tasksSKILL.md
The same SKILL.md was published earlier as sohaibdevv/claude-starter-kit/tasks.
/tasks — Task Management (SQLite-backed)
First: Read LEARNINGS.md (in this skill's directory) before proceeding.
Primary store: tasks.db (in workspace root). Markdown export at state/backlog.md (read-only, regenerated on changes).
Arguments: $ARGUMENTS
Routing
| Input | Action | |
|---|---|---|
(empty) or status | Show Status | |
add <description> | Add Task | |
review | Review & Reprioritize | |
done T### | Complete Task | |
metrics | Show Metrics |
Show Status
Flag:
- Any P1 task not
in_progress-- prompt to start it - Any task
in_progressfor 7+ days (check updated_at) -- suggest reviewing
Add Task
- Parse
<description>from arguments - Get next ID:
sqlite3 ~/claude-assistant/tasks.db "SELECT value FROM config WHERE key='next_id'" - Read
knowledge/problems/00-overview.mdto know the user's problems - Ask: which problems (by number) does this relate to? Effort (S/M/L)? Which subgoal?
- Evaluate priority:
- Problems >= 3 -- strong P1 candidate
- Problems >= 2 + urgency -- P2
- Problems == 1 or no urgency -- P3
- Problems == 0 -- flag: "Is this drift?"
- On approval, insert via sqlite3:
- Increment next_id:
sqlite3 ~/claude-assistant/tasks.db "UPDATE config SET value='T###' WHERE key='next_id'" - Re-export:
python3 ~/claude-assistant/scripts/db.py export
Review & Reprioritize
- Show all active tasks
- Read
knowledge/problems/00-overview.mdandknowledge/user/goals.md - For each task: priority still correct? Stale? Blocking others? Connected to active subgoal?
- Present proposed changes:
| ID | Current P | Proposed P | Reason | - On approval, update and re-export
Complete Task
- Parse task ID
- Update:
- Re-export:
python3 ~/claude-assistant/scripts/db.py export - Show: task name, problems addressed, days from created to done
- Suggest next task based on priority
Show Metrics
Task Conventions
- IDs:
T+ 3-digit counter (T001, T002, ...), monotonic, never reused - Priority: P1 (now), P2 (this week), P3 (this month), BL (backlog)
- Effort: S (< 1 session), M (2-3 sessions), L (4+ sessions)
- Status: pending, in_progress, done, blocked
- Problems: comma-separated numbers referencing the user's 12 problems
- Subgoal: S1, S2, etc. from goals.md
Before Substantive Work
When the user requests work that will take more than ~5 minutes:
- Check active tasks -- does this map to one?
- If NO -- flag: "This doesn't map to an active task. Want to add it, or continue anyway?"
- If the work touches zero problems -- flag: "This isn't connected to any of your problems. Proceed?"
Security
- Security scanPassed, 100 of 100Pattern checks on the skill's files.
- Model reviewNot runA model review is scheduled for the next rescan.
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/tasks/SKILL.md
Popularity
- Repository stars
- 109
- Repository forks
- 18
- 7-day trend score
- 0
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/mp-web3/claude-starter-kit/tasks)