Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Autonomously plans and specifies system features one at a time, starting from a single user command as the immutable foundation. It locks the system domain after initialization and runs a planning loop without further user input until stopped. Each iteration produces exactly one feature specification with scope, implementation steps, and a verifiable success condition.
Eliminates repeated prompting by running a structured planning loop autonomously while enforcing strict scope boundaries to prevent feature creep.
When to use it
- Drafting a feature roadmap for a new API after describing it once
- Expanding a CLI tool design iteratively from an initial command
- Generating modular specs for a workflow automation system
- Planning components for a backend service from a single description
- Building out a product specification without repeated prompting
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: use autonomous-feature-planner to expand my task management API
- 1Captures 'task management API' as the immutable foundation command
- 2Derives and locks the system domain as a REST API for managing tasks and related entities
- 3Declares minimal assumptions required to define the domain
- 4Selects the first functionally distinct feature within the domain
- 5Outputs a labeled spec with scope, ordered implementation steps, inputs, outputs, dependencies, and one success condition
Feature 1 specification: task creation endpoint with inclusion/exclusion list, linear implementation plan, required inputs, produced outputs, and a verifiable success condition