An intelligent workflow management skill that organizes complex conversations into weighted threads to maintain focus on primary objectives.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install openclaw-task-weight-manager
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install openclaw-task-weight-manager using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Task Weight Manager?
The Task Weight Manager is a sophisticated control layer for Openclaw Skills designed to handle high-context, multi-threaded conversations. It prevents the common issue of context drift by treating a chat not as a flat transcript, but as a structured queue of weighted intents. By identifying a mainline mission and protecting it from tangential interruptions, it ensures the agent remains productive on high-value tasks while systematically archiving side topics for later attention.
This skill is particularly effective for developers and power users who often juggle multiple sub-tasks within a single session. It implements a robust weighting system based on urgency and dependency, allowing the agent to provide compact status updates and maintain a persistent thread board in the workspace. It brings order to complex reasoning paths and ensures that important side-notes are never lost in the shuffle.
Task Weight Manager Use Cases
- Managing multiple interleaved development tasks within a single long-running session.
- Protecting the primary mission from being derailed by exploratory questions or casual tangents.
- Prioritizing sub-tasks based on time sensitivity or dependencies between different work streams.
- Creating a persistent 'Parking Lot' for ideas that need to be addressed later without interrupting the current flow.
- Re-orienting the agent after a long break or complex context switch using a structured reflection loop.
How Task Weight Manager Works
- The skill analyzes recent conversation turns to split user intent into distinct, named threads with unique IDs.
- It applies weighting rules to each thread, adjusting scores based on explicit user priority, time sensitivity, and task dependencies.
- A mainline thread is established to serve as the primary focus, triggering a 'Focus Lock' that protects the current work from being abandoned.
- When interruptions occur, the skill classifies them and either addresses them immediately if urgent or parks them in a background thread.
- Through a periodic reflection loop, it re-evaluates the thread board, merges duplicates, and steers the agent back to the highest-weight task if drift is detected.
Task Weight Manager Setup
To utilize this skill, ensure your environment meets the Python 3 requirement. It is compatible across macOS, Linux, and Windows environments.
# Standard installation via Openclaw CLI
openclaw install task_weight_manager
# Or manual setup via the repository
git clone https://github.com/LucasZH7/openclaw-task-weight-manager
Enable the skill within your agent configuration to start managing Openclaw Skills threads automatically.
Task Weight Manager Data Schema & Taxonomy
The skill maintains state within the workspace to ensure continuity across sessions. Data is organized in a human-readable markdown file located at task-weight-manager/threads.md with the following structure:
| Attribute | Type | Description |
|---|---|---|
id |
String | A short label for the thread (e.g., A, B, F). |
title |
String | A concise one-line summary of the thread purpose. |
goal |
String | The specific outcome desired for this task. |
weight |
Integer | Priority score from 1-100 based on urgency and impact. |
state |
Enum | Status: mainline, active, parked, blocked, or done. |
next_action |
String | The immediate next step required to progress the thread. |
evidence |
String | Context from recent turns justifying the current status. |
Task Weight Manager Advanced Features
- Focus Lock Mechanism: Automatically acknowledges side topics while explicitly returning the conversation to the mainline task.
- Dynamic Re-prioritization: A reflection loop that re-scores tasks whenever the user requests a priority shift or during periodic system heartbeats.
- Contextual Weighting: Signals like '+40 for explicit priority' or '-20 for blocked status' ensure the prioritization is logical and transparent.
- Multi-State Persistence: Tracks threads through various lifecycles, including 'Blocked' and 'Done' states, saved directly to your workspace.
- Compact Status Blocks: Provides brief, non-intrusive UI updates showing current priority rankings (e.g., F(92) > B(61)) before proceeding with work.
SKILL.md
Loading
Related Openclaw Skills
Prompt Transformer
A specialized tool that converts raw, casual requests into highly structured and effective prompts for coding, planning, and prototyping.
GitCode PR Audit
An automated auditing tool that samples merged pull requests across multiple repositories to evaluate compliance, code quality, and documentation standards.
OpenFun
OpenFun is an AI-powered video factory that reverse-engineers viral patterns to create original short-form content for social media.
Content Idea Generator
A systematic content strategy engine that generates high-quality, validated topics using 8 creative frameworks to ensure you never run out of inspiration.
Stock Screener AI
A powerful AI-driven tool for screening A-shares and Hong Kong stocks using technical, fundamental, and capital flow indicators.
KnowAir Air Quality
A professional-grade air quality forecasting tool that provides station-based AQI data and health recommendations via the Caiyun Weather API.
Featured*
