Design custom workflow cycles for any domain to transform repetitive multi-step processes into structured, persistent agent skills.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install cycle
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 cycle 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 Cycle?
Cycle is a specialized framework within the Openclaw Skills ecosystem designed to create structured workflows for specific objectives. Unlike standard loops that repeat a single action, Cycle defines a sequence of distinct phases—such as planning, execution, and verification—to ensure high-quality, repeatable outcomes. By formalizing these workflows, users can turn manual routines into reliable automated systems.
This skill enables developers to architect the lifecycle of a task from initial request to final delivery. It focuses on the logical progression of work, ensuring that each phase meets specific validation criteria before moving to the next. By using Cycle in conjunction with other Openclaw Skills, agents can handle complex, multi-layered projects with the same consistency as a human expert.
Cycle Use Cases
- Automating software development lifecycles that require sequential planning, coding, and testing.
- Streamlining research projects that move from initial questions to data synthesis and final reporting.
- Standardizing content marketing workflows including drafting, critiquing, and publishing phases.
- Optimizing any business process where users find themselves repeating the same 3-6 steps manually.
How Cycle Works
- Identify the specific goal and the final output the cycle is intended to produce.
- Map the current manual workflow to identify inefficiencies and logical transition points.
- Break the process into 3-6 sequential phases, defining clear input and output requirements for each.
- Establish handoff protocols to determine if transitions between phases are automatic, gated by user approval, or conditional.
- Integrate memory components to capture user preferences and recurring patterns, improving the intelligence of the Openclaw Skills over time.
Cycle Setup
To implement a Cycle within your agent environment, follow the structured design process:
- Review the
discovery.mdfile to identify the user's specific workflow needs. - Create a new skill directory for your cycle:
mkdir -p skills/custom-workflow-cycle
touch skills/custom-workflow-cycle/SKILL.md
- Define the phases and handoff logic in your implementation file.
- Test the cycle by triggering the first phase and validating the output at each transition point.
Cycle Data Schema & Taxonomy
The Cycle skill organizes data through a sequential phase-gate model to ensure workflow integrity.
| Data Element | Description |
|---|---|
| Phase Definition | The unique logic, input requirements, and expected output for a single step. |
| Validation Logic | The criteria that must be met before the cycle can progress to the next phase. |
| Handoff Type | Defines if the transition is automatic, user-gated, or conditional. |
| Persistent Memory | Stores historical context, preferences, and patterns specific to the Openclaw Skills instance. |
Cycle Advanced Features
- Sub-agent delegation: Assign specific phases of a cycle to specialized sub-agents for better task focus.
- Hybrid Cycle-Loop patterns: Incorporate loops within a phase to iterate on a specific action until success criteria are met.
- Conditional branching: Dynamically change the cycle path based on the output of a previous phase.
- Knowledge persistence: Use the memory layer to avoid repeating past mistakes and refine the cycle's output over multiple runs.
SKILL.md
Loading
Related Openclaw Skills
CXO (Chief Experience Officer)
A strategic AI agent skill designed to act as a virtual Chief Experience Officer, managing customer journey mapping, service design, and experience metrics.
CV Assistant
A specialized tool for generating high-impact, ATS-optimized resumes and CVs that emphasize quantifiable professional achievements.
Customer Support
A professional AI agent framework for resolving customer issues with empathy, efficiency, and structured problem-solving.
CSV Processor
A robust utility for parsing and generating RFC 4180 compliant CSV files that ensures cross-tool compatibility.
Cycling Performance & Training
A comprehensive technical knowledge base for optimizing cycling performance, mechanical maintenance, and endurance training strategy.
Cypress
A high-performance testing skill for writing resilient E2E and component tests while eliminating flakiness and race conditions.
Featured*
