A comprehensive AI development framework that automates the software lifecycle from discovery to execution with built-in project memory.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install plan-flow
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 plan-flow 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 Plan-Flow?
Plan-Flow is a high-performance skill set designed for professional AI-assisted software development, focusing on structured workflows and persistent project intelligence. By integrating directly into the developer's workspace, it transforms fragmented coding prompts into a cohesive, documented engineering process. As a standout among Openclaw Skills, it ensures that every task follows a logical progression from requirements gathering to verified execution.
The core philosophy of Plan-Flow centers on "Flow," an automated state where the AI agent manages discovery documents, implementation plans with complexity scoring, and automated code reviews. Its most unique feature is the Project Ledger, a persistent learning journal that silently captures project-specific knowledge, recurring mistakes, and architecture patterns to ensure the agent's performance improves over the lifetime of the repository.
Plan-Flow Use Cases
- Formalizing requirements for new features using structured discovery documents.
- Generating multi-phase implementation plans with granular complexity scoring (0-10).
- Automating the execution of complex development tasks with step-by-step verification.
- Performing automated code reviews on local changes or remote GitHub Pull Requests.
- Maintaining a long-term project memory to prevent the AI from repeating past mistakes.
- Creating API integration contracts directly from documentation to guide frontend-backend development.
How Plan-Flow Works
- Run the /setup command to analyze the project patterns and initialize necessary configuration files.
- Initiate requirements gathering with /discovery to ensure all technical constraints are documented.
- Trigger /create-plan, which utilizes the discovery document to build a phased implementation roadmap.
- Execute development via /execute-plan, where the agent writes code and verifies each phase sequentially.
- Audit the resulting work using /review-code or /review-pr to ensure adherence to project standards.
- Use the /flow command to toggle autopilot mode, allowing the agent to chain these steps together with minimal user intervention.
Plan-Flow Setup
To add this to your collection of Openclaw Skills, install via Clawhub:
clawhub install plan-flow
Alternatively, clone the repository manually into your local skills directory:
git clone https://github.com/brunoscardoso/plan-flow.git ~/.openclaw/skills/plan-flow
Configure your AI provider by creating a .plan-flow.yml file in your project root:
ai:
provider: claude
anthropic_api_key: your-api-key-here
Plan-Flow Data Schema & Taxonomy
Plan-Flow maintains a strict organizational structure within the flow/ directory to manage metadata and artifacts:
| Path | Description |
|---|---|
flow/discovery/ |
Contains markdown documents for requirements gathering. |
flow/plans/ |
Stores active implementation roadmaps and phase checklists. |
flow/ledger.md |
A persistent journal for project-specific learning and memory. |
flow/contracts/ |
Holds integration contracts and API specifications. |
flow/archive/ |
Location for completed or superseded plans and discovery docs. |
flow/reviewed-code/ |
Stores reports from local code review sessions. |
Plan-Flow Advanced Features
- Autopilot Flow: Enabling /flow allows the agent to move from discovery to execution without asking for permission at every step.
- Complexity Scoring: Every plan phase is assigned a 0-10 score to help developers assess risk and implementation effort.
- GitHub Integration: Native support for the
ghCLI allows for seamless AI-driven Pull Request reviews. - Persistent Project Ledger: A cross-session memory file that ensures the agent learns from project-specific patterns and previous errors.
- Hard-Block Discovery: A safety mechanism that prevents implementation planning until a valid discovery document is generated.
SKILL.md
Loading
Related Openclaw Skills
WebMCP
WebMCP enables AI agents to interact with websites through structured JavaScript tools instead of fragile DOM scraping.
Startup Info — Portable Prompt
An AI-powered research agent that generates deep-dive investor briefings and competitive analysis for startups and private companies.
Hide OpenClaw Banner
A utility configuration to disable the automated banner message shown during OpenClaw CLI startup.
Stegstr
A steganographic tool for embedding and decoding hidden Nostr messages and arbitrary data within PNG images.
CoinMarketCap Cryptocurrency API Skill
A technical reference skill for integrating CoinMarketCap cryptocurrency data, including real-time quotes, market listings, and historical OHLCV data.
CoinMarketCap DEX API
A comprehensive API reference for accessing real-time decentralized exchange data, including token prices, liquidity pools, and security analysis.
METADATA
Featured*
