map-workflow

Map Workflow

Generate a workflow graph for a multi-agent task using one of four patterns. Present the graph to the user for approval before any agent is dispatched.

Steps

1. Parse arguments

Extract from $ARGUMENTS:

Flag	Required	Default	Description
--pattern	yes	—	Sequential, Parallel, Hierarchical, or Iterative
--agents	yes	—	Comma-separated agent names in execution order
--task	yes	—	One-sentence description of what this workflow delivers
--max-iter	no	3	Max loop iterations (Iterative pattern only)
--save	no	none	If set, write the graph to .tasks/[filename].workflow.md

If --pattern or --agents or --task is missing, print usage and stop:

Usage: /map-workflow --pattern <Sequential|Parallel|Hierarchical|Iterative> \
                     --agents "<agent1,agent2,...>" \
                     --task "<what this workflow delivers>" \
                     [--max-iter <N>] [--save <filename>]

Patterns:
  Sequential   — linear pipeline, one agent after another
  Parallel     — fan-out to multiple agents simultaneously, then merge
  Hierarchical — supervisor delegates to specialists
  Iterative    — test-driven loop until quality gate passes

Reference: docs/templates/workflow-graph.md

2. Validate agents

For each agent name in --agents, verify it exists in .claude/agents/. If an agent is not found, warn but continue:

⚠️  Agent "foo-developer" not found in .claude/agents/ — check spelling.
    Proceeding with remaining agents.

3. Generate the workflow graph

Using the pattern and agents, fill in the workflow schema from docs/templates/workflow-graph.md. Apply these rules per pattern:

Sequential: Each agent becomes one node. on_pass chains to the next agent. on_fail: stop for all nodes.

Parallel: First agent in list becomes the orchestrate node. Last agent becomes the merge node. All middle agents run in parallel from orchestrate to merge. A qa-engineer node is appended after merge if not already in the list.