bivvy list

Lists all available steps and workflows in your configuration.

Usage

bivvy list

bivvy list <workflow>

bivvy list --all

bivvy list --steps-only

bivvy list --workflows-only

bivvy list --json

bivvy list --env ci

bivvy list <workflow-name>

bivvy list --all

Example Output

  Environment: development (fallback)

  Steps:
    install (template: yarn-install)
    build — npm run build
      Compiles the project
      └── depends on: install
    deploy — bin/deploy
      └── depends on: install, build

  Workflows:
    default: install → build → deploy
      Full development setup

When using --env ci, steps restricted to other environments are shown as skipped:

  Environment: ci (--env flag)

  Steps:
    dev_only (skipped in ci)
    always_run — echo always

Arguments

Argument	Description
<workflow>	Optional. When given, only that workflow is parsed (from .bivvy/workflows/<name>.yml) and shown alongside its bundled steps. Without a target, Bivvy lists everything from discovery + headers without deep-merging.

Flags

Flag	Description
--all	Show every step and workflow from the fully merged configuration (legacy behavior). Without this flag, output is built from filesystem discovery and lightweight workflow headers, and does not deep-merge.
--steps-only	Show only the steps section
--workflows-only	Show only the workflows section
--json	Output as JSON instead of styled text
--env <ENV>	Target environment (e.g., development, ci, staging)

Discovery vs Full Merge

bivvy list defaults to a fast discovery-based view of your configuration:

• Step names come from .bivvy/config.yml plus the filename stems under .bivvy/steps/.
• Workflow names come from .bivvy/config.yml plus the filenames under .bivvy/workflows/. Each workflow file’s description and step list are read from a lightweight header — the file is not fully parsed against the schema.
• ~/.bivvy/, remote extends: URLs, and .bivvy/config.local.yml are not loaded.

Pass --all to opt into the full merged view: every file the loader can find is parsed and merged together, exactly like bivvy run sees it. Use this when you want to inspect the final merged result, including overrides from config.local.yml or user-global config.

Passing a positional <workflow> switches to a per-workflow detail view: the project file plus the named workflow file are loaded together, and other workflow files are skipped.

Output Format

Steps

Each step shows:

• Step name with template reference (e.g., step_name (template: xxx)) or inline command (e.g., step_name — command_text)
• Description or title on an indented line below
• Dependencies shown as └── depends on: dep1, dep2 on an indented line
• Environment constraints: steps restricted to other environments appear as step_name (skipped in <env>)

Workflows

Each workflow shows:

• Workflow name followed by step names joined with Unicode arrows: workflow_name: step1 → step2 → step3
• Description on an indented line below (if present)

JSON Output

When using --json, output is a JSON object with environment, steps, and workflows fields. The --steps-only and --workflows-only flags control which sections are included. Steps that are excluded by the active environment include "skipped": true.

{
  "environment": "development",
  "steps": [
    {
      "name": "ci_only",
      "command": "echo ci",
      "title": "CI only",
      "skipped": true
    },
    {
      "name": "install",
      "template": "yarn-install"
    },
    {
      "name": "build",
      "command": "npm run build",
      "title": "Compiles the project",
      "depends_on": ["install"]
    }
  ],
  "workflows": [
    {
      "name": "default",
      "steps": ["install", "build"],
      "description": "Full development setup"
    }
  ]
}