A structured problem-solving skill that forces AI agents to design detailed blueprints before executing complex, multi-step tasks.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install plan-first
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 plan-first using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
Plan First is an advanced reasoning skill based on the Plan-and-Solve research paradigm. It ensures that generative AI models behave like architects rather than builders, anticipating issues before they arise. By requiring a comprehensive plan before any action is taken, this skill significantly improves success rates for high-stakes tasks where dependencies exist between steps.
Using Openclaw Skills like Plan First allows developers to mitigate the risks of model hallucinations or mid-project logic failures. It transforms the AI from a reactive tool into a proactive problem-solver that validates its own approach against specific success criteria before committing to execution.
To use the Plan First skill within your agent environment, ensure the skill definition is accessible to your Openclaw Skills workflow. Configure your agent to trigger the planning phase for any task identified as having multiple dependencies. You can initialize a planning session using the standard template:
# Example of initializing a task with the planning skill
openclaw-agent run --skill plan-first "Implement a secure JWT authentication flow"
The skill organizes information using a structured Markdown schema to ensure human-readability and machine-parseability. Key components include:
| Section | Description | Metadata |
|---|---|---|
| Task | Clear problem statement | String |
| Plan | Ordered list of steps | Array |
| Dependencies | Prerequisites for steps | List |
| Verification | Pass/Fail criteria | String |
| Risk Analysis | Potential blockers | List |
| Rollback | Reversion strategy | String |
Loading
An intelligent learning system that identifies operational patterns and generates optimized templates to enhance agent performance.

A specialized research tool that retrieves the complete trade show exhibition history for any company using the Lensmor API.

An infinite, self-driven research agent that autonomously identifies topics, conducts deep investigations, and documents findings in Feishu.

An automated speech-to-text skill for QQ Bots that decodes Silk V3 audio and transcribes it using OpenAI Whisper.

Task Director is a sophisticated workflow management skill that treats complex AI agent tasks like movie storyboards, ensuring planning happens before execution.

A comprehensive automation skill for managing Alibaba Cloud Edge Security Acceleration (ESA) services including static deployments, serverless edge functions, and distributed storage.








































