A specialized skill for managing long-term project state through a structured, file-based workspace that survives chat resets.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install long-project-manager
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install long-project-manager using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
The Long Project Manager is a high-utility addition to the library of Openclaw Skills designed to solve the challenge of context loss during multi-session or complex AI tasks. Instead of relying solely on transient chat history, this skill establishes a durable, file-based source of truth within a dedicated project folder. This approach ensures that critical project goals, architectural decisions, and next steps remain accessible even if a session is interrupted or the context window is reset.
By leveraging a standardized set of Markdown templates, the skill enables AI agents to track progress with high precision. It is particularly effective for engineering, research, and long-form writing where maintaining a clear 'judgment' and 'next action' is vital for continuity. This system transforms the AI's workspace into a professional-grade project environment, ensuring that no discovery or decision is lost to the ephemeral nature of standard chat interfaces.
To implement this as part of your Openclaw Skills setup, ensure the project templates are available in your assets directory. The skill operates by managing files within a local directory structure.
# Ensure the project directory exists
mkdir -p projects/<your-project-name>
# The agent will typically automate the template copy process
cp -r assets/project-template-en/* projects/<your-project-name>/
No additional external APIs are required, as the skill relies on standard file system operations to maintain state.
The skill organizes project data into a structured hierarchy within the projects/<name>/ directory using the following schema:
| File Name | Data Description |
|---|---|
| README.md | Core project goals, success criteria, and entry points. |
| STATUS.md | Current goal, active judgment, and immediate next action. |
| TODO.md | A prioritized list of tasks and open loops. |
| DECISIONS.md | Documentation of significant choices and technical trade-offs. |
| LOG.md | Dated entries of discoveries, command outputs, and work progress. |
| REFERENCES.md | Repository of links, file paths, and essential commands. |
| HANDOFF.md | A compact resume capsule for session transition. |
Loading
A powerful AI video generation skill that leverages the Pixwith Veo 3.1 model for high-fidelity text-to-video and image-to-video workflows.

A powerful CLI-based tool for local order lifecycle management and sales reporting.

An AI-driven automation tool designed to analyze project requirements and generate high-conversion freelance proposals for major platforms.

A comprehensive guided workflow that assists users in creating high-quality voice clones by managing recording standards, audio quality analysis, and first-run synthesis.

A technical framework for coordinating multiple AI models through quality control loops, task routing, and cost-optimized fallback chains.

Vuact is a powerful compatibility layer that enables seamless two-way interoperability between React and Vue 3 components.








































