Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
This repository provides a suite of tools designed for AI agent engineers to make codebases legible and executable. It focuses on the infrastructure required for coordinated CLI-agent work, from establishing agent-ready scaffolds to managing race-safe delegation.
The skills enable the standardization of reproducible environments, allowing agents to operate within isolated cloud boxes or localized stacks while maintaining verifiable end-to-end test suites.
WHO IT'S FOR
AI Agent Engineers
build verifiable multi-agent development harnesses
Platform Engineering Teams
standardize reproducible local and cloud environments
Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.
Follow the documented setup, then try a first task.
Claude Code marketplace bundle
Installs: Complete AI Builder Club “skills” plugin bundle; skillPaths is a bounded list of the supplied, positively mapped skills, not the bundle's complete inventory. · Claude Code
Before you start
Claude Code with plugin marketplace support
Enter these commands sequentially in a Claude Code agent session to add the AI Builder Club marketplace and install its complete skills plugin.
Replace <REPOSITORY_PATH> with the code repository to inspect.
Provide access to the repository's configuration and environment-example files; do not include secrets.
Use the dev-local-setup skill to inspect the repository at <REPOSITORY_PATH>. For this first pass, do not start servers or generate the launcher yet. Produce the small service → exact command → port → dependency table described by the skill, and list any missing facts that must be resolved before creating scripts/dev-local.sh.
Replace <SOURCE_MATERIAL_PATH> with an article, workflow note, architecture sketch, or process description.
Replace <OUTPUT_DIRECTORY> with the desired output directory.
Python 3.10+ and Pillow 10+ are required for rendering.
Use the visual-flow-gif skill to turn <SOURCE_MATERIAL_PATH> into a clear default-light-theme flow diagram. Save the static PNG, animated GIF, and editable JSON specification under <OUTPUT_DIRECTORY>. Preserve the source's actual actors, steps, decisions, stores, feedback loops, and arrow directions; do not invent missing structure. Inspect the PNG and GIF and run the documented verification and checks before delivery.
Replace <SITE_OR_CLUSTER> with the site or topic cluster being evaluated.
Replace <SEARCH_CONSOLE_SUMMARY> with its current rankings, clicks per month, and any relevant Google Search Console rows; state explicitly if no data exists.
Use the seo-growth skill to assess <SITE_OR_CLUSTER>. Based on <SEARCH_CONSOLE_SUMMARY>, determine whether this cluster belongs in Playbook A (cold start) or Playbook B (with data), explain the classification briefly, and recommend one small next action. Focus on where to direct SEO effort, not on writing an article.
Claude
Codex
Copilot
OpenCode
Grok