developing-cast

# Developing Backend's Cast Implement LangGraph casts following Backend Act patterns. ## When to Use - Building nodes, agents, tools, or graphs - Need LangGraph implementation patterns - Have architecture specs to implement ## When NOT to Use - Architecture design → `architecting-act` - DeepAgent harness (create_deep_agent, backends, subagents) → `developing-deepagent` - Project / cast scaffolding → run `uv run act new` (project) or `uv run act cast` (new cast) directly - Testing → `testing-cast` --- ## Implementation Workflow ### Step 1: Understand CLAUDE.md > **Act** = Project, **Cast** = Graph/Workflow in the project **If CLAUDE.md exists:** 1. Read `/CLAUDE.md` → Act overview, find target cast 2. Read `/casts/{cast_slug}/CLAUDE.md` → Architecture diagram, node specs 3. Proceed to Step 2 **If CLAUDE.md not found:** AskUserQuestion Format: ```json { "question": "CLAUDE.md not found. Create architecture first?", "header": "Architecture", "multiSelect": false, "options": [ {"label": "Yes", "description": "Switch to architecting-act skill"}, {"label": "No", "description": "Proceed without architecture specs"} ] } ``` - Yes → use `architecting-act` skill first - No → proceed to Step 2 ### Step 2: Implementation **Implement in order:** state → dependency modules → nodes → conditions → graph - Use Component Reference tables below for each component type ``` 1. State (state.py) # Foundation ↓ 2. Dependency modules # agents, m...