AnsibleClaw is a Python-based framework designed to bridge the gap between Ansible's extensive library of modules and the capabilities of AI agents, such as Cursor and Claude Code. Its primary function is to convert Ansible modules into portable AI agent skill packages, enabling these agents to leverage Ansible's automation power. The project aims to empower AI agents to perform infrastructure management, configuration, and deployment tasks by interacting with the Ansible CLI.

The core functionality of AnsibleClaw revolves around generating complete skill packages from Ansible module documentation obtained through `ansible-doc`. Each generated package is self-contained and includes a `SKILL.md` file, wrapper scripts, prerequisite checks, and a ready-to-use Ansible playbook. The `SKILL.md` file provides the AI agent with instructions on how to use the module, including parameter descriptions, usage examples, and safety guidelines. The wrapper script (`run.sh`) executes the Ansible CLI commands, while the prerequisite check script (`check.sh`) verifies that Ansible and the required module are available. The playbook (`playbook.yml`) provides a basic example of how to use the module. A key advantage of AnsibleClaw is that the generated skills only require `ansible-core` at runtime, eliminating the need for custom packages or dependencies beyond the standard Ansible installation.

The process of using AnsibleClaw involves three main steps: searching for the desired Ansible module, generating a skill package, and then using the generated skill within an AI agent. The `ansibleclaw search` command allows users to find modules by keyword or namespace. Once a module is identified, the `ansibleclaw generate` command creates a skill package. This command offers options to install the skill directly into supported AI agents like Cursor or Claude Code, create a distributable ZIP archive, or specify a custom output directory. The generated skills are designed to be portable and can be easily copied into the skill directories of various AI agents.

AnsibleClaw also provides a web dashboard, accessible via the `ansibleclaw ui` command. The dashboard offers a user-friendly interface for managing skills, searching for modules, and generating new skills. The "Skills Library" allows users to view, download, and install generated skills. The "Module Search" feature enables users to search for modules and view their documentation. The "Skill Generator" allows users to generate skills directly from the web interface, previewing the `SKILL.md` content in real-time.

The project includes several built-in skills that enhance the functionality of AI agents. These include `ansible_manager`, which enables AI agents to execute any Ansible module via the CLI; `ansible_search`, which allows AI agents to discover modules using `ansible-doc`; and `ansible_skills_factory`, which allows AI agents to generate new skills on-demand.

The project is structured with a clear separation of concerns, including a CLI interface, core logic for parsing and packaging, built-in skills, templates for skill generation, and an optional web dashboard. The project is designed to be easily installed via pip, with options for installing the core CLI, the web dashboard, and development tools. The project is licensed under the Apache License 2.0. The comprehensive user guide, located in the `docs/user-guide.md` file, provides detailed information on CLI usage, the web dashboard, workflows, and troubleshooting.