streamline
MANDATORY PREPARATION
Invoke /agent-workflow — it contains workflow principles, anti-patterns, and the Context Gathering Protocol. Follow the protocol before proceeding — if no workflow context exists yet, you MUST run /teach-maestro first. Consult the agent-architecture reference in the agent-workflow skill for complexity assessment and topology simplification.
Reduce complexity without reducing capability. Every component should earn its place.
Streamlining Analysis
For each component, ask:
- Does this add measurable value? If you can't name the specific value, remove it.
- Can this be combined with another component? Merge overlapping responsibilities.
- Is this solving a real problem or an imagined one? Remove speculative complexity.
- Would a simpler alternative work? Prefer simplicity over elegance.
Common Streamlining Targets
Pipeline Steps
- Remove steps that transform data without changing it meaningfully
- Combine sequential steps that could be one prompt
- Eliminate validation steps that duplicate downstream validation
Tool Consolidation
- Merge tools that operate on the same data with different filters
- Remove tools the model never selects (check usage logs)
- Combine read tools with similar signatures
Prompt Simplification
- Remove instructions the model follows by default
- Consolidate redundant constraints
- Shorten few-shot examples to minimum viable length
Configuration Reduction
- Remove config parameters that always use the default
- Hardcode values that never change between environments
- Merge related config into logical groups
Streamlining Report
For each recommendation:
- What to remove/simplify — specific component or code
- Why it's safe — what ensures functionality is preserved
- Expected impact — latency reduction, cost reduction, or maintainability improvement
Complexity Score
| Component | Current | Minimal Viable | Action |
|---|---|---|---|
| Pipeline steps | ? | ? | Remove/merge ? |
| Tools | ? | ? | Consolidate ? |
| Config params | ? | ? | Remove ? |
| Agent count | ? | ? | Collapse ? |
Streamlining Checklist
- Each component justified with concrete value
- Redundant steps identified and marked for removal
- Tool usage logs checked for never-used tools
- Golden test set passes after each simplification
- No safety guardrails, logging, or error handling removed
Recommended Next Step
After streamlining, run /evaluate to verify functionality is preserved, then /refine for final polish.
NEVER:
- Remove error handling in the name of simplicity
- Streamline without verifying the golden test set still passes
- Remove logging or observability
- Simplify below the level of correctness
- Remove safety guardrails
More from sharpdeveye/maestro
agent-workflow
Use when any Maestro command is invoked — provides foundational workflow design principles across prompt engineering, context management, tool orchestration, agent architecture, feedback loops, knowledge systems, and guardrails.
133diagnose
Use when the user wants to find problems, audit workflow quality, or get a comprehensive health check on their AI workflow.
131evaluate
Use when the user wants a quality review, interaction audit, or to test the workflow against realistic scenarios.
130calibrate
Use when workflow components are inconsistent, naming conventions vary, or a new team member's work needs alignment to project standards.
125fortify
Use when the workflow lacks error handling, has been failing in production, or needs retry logic, fallback strategies, and circuit breakers.
125teach-maestro
Use when starting a new project with Maestro or when no .maestro.md context file exists yet. Run once per project.
125