maestroresearch

name

maestro:research

description

Structured research workflow for maestro features. Guides tool selection across three tiers (codebase exploration, Context7 for library docs, NotebookLM for deep analysis), defines research patterns, finding organization via memory_write, and completion criteria. Use during the research pipeline stage after feature_create and before plan_write. Also use when investigating a problem space, comparing technical approaches, gathering context on unfamiliar code, or needing to understand external library APIs before making architectural decisions.

Research Phase

Research sits between discovery and planning. Gather enough understanding to write a confident plan. Question assumptions, start with the smallest investigation that could answer the question, and scale up only when complexity warrants it.

When to Research

Enter research when:

• A feature exists (feature_create done)
• The problem involves unfamiliar code, external libraries, or multiple valid approaches
• You can't yet write a confident ## Discovery section for the plan

Skip research when the task is mechanical (rename, config change, straightforward addition) or you already have deep context from prior work.

Research Tiers

Your approach adapts to what tools are available.

Tier 1: Always Available

Codebase exploration -- Agent subagents

Use for understanding existing code, finding patterns, tracing data flow. Spawn focused subagents with one clear objective each.

memory_write({ feature: "my-feature", name: "research-existing-auth-patterns", content: "..." })

Web search -- WebSearch + WebFetch

Use for error messages, ecosystem comparisons, best practices. Save fetched content to a local file or memory immediately -- interrupted sessions lose unsaved fetches.

Tier 2: Context7

Available when: Context7 MCP server is configured.