Installing Ansible AI Skills with Lola and AI Forge
-
Gineesh Madapparambath
- Ai tools, Ansible
- July 26, 2026
In the previous post, we covered what Lola is and why it matters. Now let’s put it to work: installing Ansible community AI skills from the ai-forge project and seeing how they fit into an Ansible development workflow.
Note
This post isn’t affiliated with or endorsed by the Ansible community, ai-forge maintainers, Red Hat, or the Lola project. It’s an independent walkthrough based on the public ai-forge and Lola GitHub repositories.
What is AI Forge?
AI Forge is an Ansible community project, hosted under the ansible-community GitHub organization, that packages AI assistant skills and commands for developing Ansible content that follows Red Hat Communities of Practice (CoP) automation good practices.
Instead of manually referencing documentation or remembering conventions, ai-forge gives your AI coding assistant the knowledge to guide you through creating, reviewing, and improving Ansible content: roles, collections, modules, playbooks, and tests.
Warning
The ai-forge repository describes itself as brand new, with structure and contents that may change frequently as AI tooling and standards evolve. As with any AI-generated output, careful human review is still needed.
Available Modules
Ai-forge ships as a Lola marketplace with 7 modules. Numbers below combine the project’s auto-generated SKILLS.md index, each module’s README, and a live lola search --mod check, which is the most current source since the repo updates frequently:
| Module | Skills | Commands | What It Does |
|---|---|---|---|
| ansible-collection-standards | 1 | 3 | CoP standards compliance review, collection scaffolding, and inclusion review |
| ansible-role | 0 | 1 | Interactive role scaffolding with realistic defaults |
| ansible-collection-sdlc | 26 | 3 | Full dev lifecycle: commits, branches, PRs, releases, testing, SonarCloud |
| ansible-content-development | 4 | 0 | Content authoring: modules, playbooks, roles, Molecule scenarios |
| cloud_content | 3 | 0 | AWS terminator analysis, implementation, and PR workflow for cloud collections |
| network_content | 1 | 0 | Triage for network collections (cisco.ios, arista.eos, junos, and related) |
| ansible-documentation | 1 | 0 | Fetch and render Ansible documentation from docs.ansible.com |
Note
Module counts can drift quickly on an actively developed repo. Ai-forge auto-generates SKILLS.md from the SKILL.md files in each module, so it’s the most reliable source if you want current numbers before you install.
Step-by-Step Installation
1. Install Lola
pip install lola-ai
2. Add the Ansible Content Marketplace
Ai-forge publishes its modules through a Lola marketplace. Register it:
lola market add ansible-content \
https://raw.githubusercontent.com/ansible-community/ai-forge/main/lola-market.yml
Search the marketplace to confirm the modules are discoverable:
$ lola search --mod ansible
Local registry (5 modules)
ansible-collection-sdlc
26 skills, 3 commands, 0 agents
ansible-collection-standards
1 skill, 3 commands, 0 agents
ansible-content-development
4 skills, 0 commands, 0 agents
ansible-documentation
1 skill, 0 commands, 0 agents
ansible-role
0 skills, 1 command, 0 agents
Note
Older Lola versions used lola mod search. That form is now deprecated in favor of lola search --mod, so use whichever your installed version reports back to you.
3. Install Modules
Since Ansible work often spans multiple projects, you can install to user scope so the skills are available everywhere:
lola install ansible-collection-standards -a claude-code --scope user
lola install ansible-role -a claude-code --scope user
lola install ansible-collection-sdlc -a claude-code --scope user
lola install ansible-content-development -a claude-code --scope user
lola install cloud_content -a claude-code --scope user
lola install network_content -a claude-code --scope user
Or install to the current project only (default scope):
lola install ansible-collection-sdlc -a claude-code
4. Verify Installation
$ lola list
Installed (7 modules)
ansible-collection-standards
- scope: user
assistants: [claude-code]
ansible-role
- scope: user
assistants: [claude-code]
ansible-collection-sdlc
- scope: user
assistants: [claude-code]
ansible-content-development
- scope: user
assistants: [claude-code]
cloud_content
- scope: user
assistants: [claude-code]
network_content
- scope: user
assistants: [claude-code]
ansible-documentation
- scope: user
assistants: [claude-code]
5. Start a New Claude Code Session
Skills and commands are read when Claude Code loads its configuration, so start a fresh session after installing for the new skills and slash commands to be picked up. In Claude Code, you can check what’s available by typing /skills in the prompt.
What Gets Installed Where
For Claude Code at user scope, Lola writes to your ~/.claude/ configuration directory: skills go under ~/.claude/skills/, and commands under ~/.claude/commands/. Exactly how agent-level guidance is merged into any existing CLAUDE.md depends on the version of Lola you’re running, so check the Lola documentation for the current behavior rather than relying on this post for that detail.
How the Mechanisms Differ
Commands are invoked explicitly by typing the command name in Claude Code. For example, /ansible-cop-review (from the ansible-collection-standards module) triggers a CoP compliance review.
Skills activate automatically when your request matches the skill’s trigger conditions. Ask Claude Code to “write a module” and the write-module skill (from ansible-content-development) activates with Ansible module development guidance.
Using the Skills in Practice
Review Code Against CoP Best Practices
/ansible-cop-review
Example:
❯ /ansible-cop-review
Thought for 4s (ctrl+o to expand)
● I'll review the Ansible code in this project against Red Hat CoP automation good
practices. Let me start by discovering the project scope and finding all Ansible files.
Searched for 3 patterns (ctrl+o to expand)
...
This command (from ansible-collection-standards) reviews Ansible code against Red Hat CoP automation good practices with severity classification, a diff-aware mode, and auto-fix capabilities, per the module’s own documentation.
Scaffold a New Role
/ansible-scaffold-role
This command (from ansible-role) scaffolds a new role with an interactive variable builder, task componentization, smart handler generation, and CoP compliance.
Scaffold a New Collection
/ansible-scaffold-collection
This command (from ansible-collection-standards) scaffolds a new content collection with plugin generation, CI/CD pipelines, and CoP compliance.
Review a Collection for Community Inclusion
/ansible-collection-standards-inclusion-review
Reviews an Ansible collection for inclusion in the Ansible community package.
Write an Ansible Module
Just ask Claude Code to write a module and the write-module skill activates automatically:
Write an Ansible module for managing nginx virtual hosts
Per the ansible-content-development module’s skill description, this scaffolds new modules or reviews existing ones against official best practices from docs.ansible.com.
Create Conventional Commits
commit these changes
The commit skill (from ansible-collection-sdlc) creates conventional commits with FQCN scopes for Ansible collection content.
Create a Pull Request
create a PR for these changes
The create-pr skill runs pre-flight checks, validates changelog fragments, and creates a draft PR.
Check CI Status
/check-pr-actions
This command (from ansible-collection-sdlc) checks GitHub Actions or GitLab CI status for a pull request and analyzes failures. A related command, /check-pr-sonarcloud, checks SonarCloud analysis results.
Agent Roles in the SDLC Module
The ansible-collection-sdlc module doesn’t currently define any subagents of its own (its README lists “None currently defined” under Agents). Instead, its 26 skills cover the collection development lifecycle end to end: changelog-fragment, commit, create-branch, create-pr, pr-review, release, run-tests, sanity, security-scan, remove-deprecations, tox-lint, version helpers (current-release, next-release), stable-branch release orchestration (stable-release, stable-release-analyze, stable-release-prep), and SonarCloud integration (configure-sonarcloud-collection, configure-sonarcloud-coverage, sonarcloud-remediation).
Optional Configuration
The ansible-collection-sdlc module supports an optional configuration file for defaults like your collections path, GitHub username, and sanity test mode:
cp ansible-release.conf.template ~/.ansible-release.conf
vim ~/.ansible-release.conf
source ~/.ansible-release.conf
See the module’s ansible-release.conf.template in the repo for the full list of options.
Updating Modules
When ai-forge publishes updates:
# Update module sources
lola mod update
# Regenerate assistant files
lola update
```text
## Licensing
This is worth flagging clearly since the two projects use different licenses: the Lola tool itself is released under **Apache-2.0**, while the **ai-forge** content (the Ansible skills, commands, and modules themselves) is released under **GPL-3.0-or-later**. If you're building on either project, check the license that applies to what you're actually using.
## Tips
- Use `--scope user` if you work across multiple Ansible projects and want the skills available everywhere; use the default project scope to keep them contained to one project.
- Start a new Claude Code session after installing so new skills and commands are picked up.
- Some ai-forge modules (`ansible-content-development`, `cloud_content`, `network_content`) currently ship skills only, with no slash commands; others (`ansible-role`) ship a command only, with no auto-triggered skills.
- Optional dependencies referenced by the SDLC and standards modules include `ansible-creator`, `ansible-lint`, `ansible-test`, `antsibull-changelog`, and the `gh` CLI. None are strictly required, but some skill features fall back to manual steps without them.
## References
- [AI Forge GitHub Repository](https://github.com/ansible-community/ai-forge)
- [AI Forge Skills Index (SKILLS.md)](https://github.com/ansible-community/ai-forge/blob/main/SKILLS.md)
- [Lola GitHub Repository](https://github.com/LobsterTrap/lola)
- [Lola Documentation](https://lobstertrap.org/lola/)
- [Lola Community Marketplace](https://github.com/RedHatProductSecurity/lola-market)
- [Red Hat CoP Automation Good Practices](https://github.com/redhat-cop/automation-good-practices)
Gineesh Madapparambath
Gineesh Madapparambath is the founder of techbeatly. He is the co-author of The Kubernetes Bible, Second Edition and the author of Ansible for Real Life Automation. He has worked as a Systems Engineer, Automation Specialist, and content author. His primary focus is on Ansible Automation, Containerisation (OpenShift & Kubernetes), and Infrastructure as Code (Terraform). (Read more: iamgini.com)
Note
Disclaimer: The views expressed and the content shared in all published articles on this website are solely those of the respective authors, and they do not necessarily reflect the views of the author’s employer or the platform. We strive to ensure the accuracy and validity of the content published on our website. However, we cannot guarantee the absolute correctness or completeness of the information provided. It is the responsibility of the readers and users of this website to verify the accuracy and appropriateness of any information or opinions expressed within the articles. If you come across any content that you believe to be incorrect or invalid, please contact us immediately so that we can address the issue promptly.