Level 2: Rapid Workflows
Complexity: Low-Medium | Artifacts: Memory/Lightweight files | State: Session-scoped
Level 2 workflows provide lightweight planning or single analysis with fast iteration. They're designed for tasks with relatively clear requirements that don't need full session persistence.
Overview
Included Workflows
| Workflow | Purpose | Artifacts | Execution |
|---|---|---|---|
lite-plan | Clear requirement development | memory://plan | -> lite-execute |
lite-fix | Bug diagnosis and fix | .workflow/.lite-fix/ | -> lite-execute |
multi-cli-plan | Multi-perspective tasks | .workflow/.multi-cli-plan/ | -> lite-execute |
Common Characteristics
| Property | Value |
|---|---|
| Complexity | Low-Medium |
| State | Session-scoped / Lightweight persistence |
| Execution | Unified via lite-execute |
| Use Case | Relatively clear requirements |
Workflow 1: lite-plan -> lite-execute
In-memory planning + Direct execution
Command
/workflow:lite-plan "Add user authentication API"
/workflow:lite-execute
Flow Diagram
Process Phases
Phase 1: Code Exploration (Optional)
# If -e flag specified
/workflow:lite-plan -e "Add user authentication API"
- Multi-angle code analysis via cli-explore-agent
- Identifies patterns, dependencies, integration points
Phase 2: Complexity Assessment
- Low: Direct planning without agent
- Medium/High: Use cli-lite-planning-agent
Phase 3: Planning
- Load plan schema:
~/.claude/workflows/cli-templates/schemas/plan-json-schema.json - Generate plan.json following schema
Phase 4: Confirmation & Selection
- Display plan summary (tasks, complexity, estimated time)
- Ask user selections:
- Confirm: Allow / Modify / Cancel
- Execution: Agent / Codex / Auto
- Review: Gemini / Agent / Skip
Phase 5: Execute
- Build executionContext (plan + explorations + clarifications + selections)
- Execute via
/workflow:lite-execute --in-memory
Artifacts
- In-memory:
memory://plan(not persisted) - Optional:
.workflow/.lite-exploration/(if code exploration used)
Use Case
Clear single-module features
Example:
/workflow:lite-plan "Add email validation to user registration form"
/workflow:lite-execute
Workflow 2: lite-fix
Intelligent diagnosis + Fix (5 phases)
Command
/workflow:lite-fix "Login timeout after 30 seconds"
/workflow:lite-execute --in-memory --mode bugfix
# Emergency hotfix mode
/workflow:lite-fix --hotfix "Production database connection failing"
Flow Diagram
Process Phases
Phase 1: Bug Analysis & Diagnosis
- Intelligent severity pre-assessment (Low/Medium/High/Critical)
- Parallel cli-explore-agent diagnosis (1-4 angles)
Phase 2: Clarification (Optional)
- Aggregate clarification needs
- AskUserQuestion for missing info
Phase 3: Fix Planning
- Low/Medium severity -> Direct Claude planning
- High/Critical severity -> cli-lite-planning-agent
Phase 4: Confirmation & Selection
- Display fix-plan summary
- User confirms execution method
Phase 5: Execute
- Execute via
/workflow:lite-execute --in-memory --mode bugfix
Artifacts
Location: .workflow/.lite-fix/{bug-slug}-{YYYY-MM-DD}/
.workflow/.lite-fix/
└── login-timeout-fix-2025-02-03/
├── diagnosis-root-cause.json
├── diagnosis-impact.json
├── diagnosis-code-flow.json
├── diagnosis-similar.json
├── diagnoses-manifest.json
├── fix-plan.json
└── README.md
Severity Levels
| Severity | Description | Planning Method |
|---|---|---|
| Low | Simple fix, clear root cause | Direct Claude (optional rationale) |
| Medium | Moderate complexity, some investigation | Direct Claude (with rationale) |
| High | Complex, multiple components affected | cli-lite-planning-agent (full schema) |
| Critical | Production incident, urgent | cli-lite-planning-agent + hotfix mode |
Hotfix Mode
/workflow:lite-fix --hotfix "Production API returning 500 errors"
Characteristics:
- Skip most diagnosis phases
- Minimal planning (direct execution)
- Auto-generate follow-up tasks for complete fix + post-mortem
Use Cases
- Bug diagnosis
- Production emergencies
- Root cause analysis
Workflow 3: multi-cli-plan -> lite-execute
Multi-CLI collaborative analysis + Consensus convergence
Command
/workflow:multi-cli-plan "Compare Redis vs Memcached for caching"
/workflow:lite-execute
Flow Diagram
Process Phases
Phase 1: Context Gathering
- ACE semantic search
- Build context package
Phase 2: Multi-CLI Discussion (Iterative)
- cli-discuss-agent executes Gemini + Codex + Claude
- Cross-verification, synthesize solutions
- Loop until convergence or max rounds
Phase 3: Present Options
- Display solutions with trade-offs
Phase 4: User Decision
- User selects solution
Phase 5: Plan Generation
- cli-lite-planning-agent generates plan
- -> lite-execute
Artifacts
Location: .workflow/.multi-cli-plan/{MCP-task-slug-date}/
.workflow/.multi-cli-plan/
└── redis-vs-memcached-2025-02-03/
├── context-package.json
├── rounds/
│ ├── round-1/
│ │ ├── gemini-analysis.md
│ │ ├── codex-analysis.md
│ │ ├── claude-analysis.md
│ │ └── synthesis.json
│ ├── round-2/
│ └── ...
├── selected-solution.json
├── IMPL_PLAN.md
└── plan.json
vs lite-plan Comparison
| Aspect | multi-cli-plan | lite-plan |
|---|---|---|
| Context | ACE semantic search | Manual file patterns |
| Analysis | Multi-CLI cross-verification | Single planning |
| Iteration | Multiple rounds until convergence | Single round |
| Confidence | High (consensus-driven) | Medium (single perspective) |
| Time | Longer (multi-round) | Faster |
Use Cases
- Multi-perspective analysis
- Technology selection
- Solution comparison
- Architecture decisions
Level 2 Comparison Table
| Aspect | lite-plan | lite-fix | multi-cli-plan |
|---|---|---|---|
| Purpose | Clear requirements | Bug diagnosis | Multi-perspective |
| Planning | In-memory | Severity-based | Consensus-driven |
| Artifacts | memory://plan | .lite-fix/ | .multi-cli-plan/ |
| Code Exploration | Optional (-e flag) | Built-in diagnosis | ACE search |
| Multi-CLI | No | No | Yes (Gemini/Codex/Claude) |
| Best For | Single-module features | Bug fixes | Technology decisions |
Execution: lite-execute
All Level 2 workflows execute via lite-execute:
/workflow:lite-execute --in-memory
Execution Flow
Features
- Parallel execution for independent tasks
- Sequential phases for dependent tasks
- Progress tracking via TodoWrite
- Optional code review
Related Workflows
- Level 1: Ultra-Lightweight - Simpler workflow
- Level 3: Standard - Full session management
- Level 4: Brainstorm - Multi-role exploration
- FAQ - Common questions