initiative

Plans and executes multi-task development initiatives with shared memory and per-task handoffs.

54|6|Updated Jan 7, 2026
One-click install
npx skills add https://github.com/schmitthub/clawker --skill initiative-schmitthub
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: initiative
Source: https://github.com/schmitthub/clawker/tree/main/.agents/skills/initiative
Command: npx skills add https://github.com/schmitthub/clawker --skill initiative-schmitthub

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes assets (resource) components.

What problem does it solve? Large development efforts that span multiple conversations lose context between sessions, causing duplicated work, forgotten dependencies, and untracked progress. This Skill structures that work into a shared plan with tracked tasks, acceptance checks, and clean handoffs between conversations. ## Core Features & Use Cases - Shared planning memory: Creates a structured initiative plan in Serena memory with objectives, scope, task dependencies, and measurable completion criteria. - One task per conversation: Executes a single task per session, runs its acceptance checks, records results, and stops with a self-contained handoff prompt for the next conversation. - Status tracking and reviews: Tracks tasks as pending, in_progress, blocked, or complete, and runs code reviews before marking work done. - Use Case: You are refactoring a large Go codebase across five phases. Create an initiative plan, then complete one phase per conversation, with each session resuming exactly where the last left off using the recorded memory and handoff prompt. ## Quick Start Ask the agent to create an initiative plan for your multi-step development work, or name an existing initiative memory and task to continue it.

Frequently Asked Questions about initiative

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I split a large coding project across multiple AI conversations?▼

Create an initiative plan that divides the work into tasks small enough for one conversation each, with dependencies and acceptance checks. Each session completes one task, records results in shared memory, and produces a handoff prompt for the next conversation.

How do I resume unfinished work in a new conversation?▼

Name the initiative memory and task, or let the skill select the first pending task whose dependencies are complete. It reads the recorded progress, checks the current branch and working tree, and resumes from the tracked state.

Does this skill work without Serena memory tools?▼

Yes. When Serena memory tools are unavailable, it reads and writes the same memory files directly under .serena/memories/, appending .md to the memory name.

When should I not use a multi-task initiative plan?▼

Avoid it for routine single-task changes that fit in one conversation. The planning, tracking, and handoff overhead only pays off when work genuinely spans multiple sessions with task dependencies.

Can the skill commit changes automatically after each task?▼

No. It commits only when the current session explicitly authorizes it, including only that task's changes. The skill grants no extra permission to commit, push, or modify external systems.