Use this skill when you need to author, run, inspect, or debug a dynamic agent workflow using Open Agent Workflows (owf). Reach for it whenever a task requires orchestrating multiple LLM calls — fan-out research, staged pipelines, multi-provider review, schema-validated extraction — with durability, resumability, and offline testability. The runtime is stdlib-only Python (>=3.11), works fully offline with the fake provider, and shells out to the claude or codex CLIs for real model calls.
$npx -y skills add akakabrian/agent-workflows --skill agent-workflowsInstalls into the current project.
Run `npx skills use "https://github.com/akakabrian/agent-workflows" --skill "akakabrian/agent-workflows"` and follow the generated skill instructions now. Read its complete output, redirecting it to a temporary file first if necessary. Resolve relative paths from the supporting-files directory it provides.
Use the skills in "https://github.com/akakabrian/agent-workflows" that are relevant to the current task. Run `npx skills add "https://github.com/akakabrian/agent-workflows"` and select the relevant skills, then follow their instructions.