| name | init |
| description | Copy a bundled flowai-workflow into the user's project so they can adapt and run it locally. Use when the user wants to start using flowai-workflow in a new repo. |
| argument-hint | workflow name (default github-inbox) or --list |
| effort | low |
Init a flowai-workflow into the user's project
Copy one of the engine-bundled workflows into the user's current
project at <project-root>/.flowai-workflow/<name>/. Adaptation
(project-specific commands, paths, prompts) happens after copy via
the scaffold skill — init itself is a verbatim copy.
The flowai-workflow binary is a plugin precondition (FR-E78) and
embeds the canonical workflow catalogue at compile time.
Preflight
command -v flowai-workflow >/dev/null 2>&1 || { echo "Error: flowai-workflow is required — see https://github.com/korchasa/flowai-workflow#install" >&2; exit 127; }
List bundled workflows
flowai-workflow init --list
This enumerates the workflows shipped in the binary. Pick one by
intent (issue-driven SDLC → github-inbox; autonomous local SDLC →
autonomous-sdlc; smoke check → github-inbox-opencode-test).
Scaffold into the project
flowai-workflow init --workflow <name>
Add --dry-run first when the user wants to preview writes.
The scaffold lands under <cwd>/.flowai-workflow/<name>/. After it
completes, hand off to the scaffold skill to adapt placeholders and
project-specific commands.
Follow-up
init is a copy operation, not a completion. Always remind the user:
- The copied workflow needs project-specific adaptation (commands,
secrets, paths) — run
/flowai-workflow:scaffold <name> next.
- To execute the workflow, use
/flowai-workflow:run .flowai-workflow/<name> from the project root.