search-planning

Generate a traceable multi-step research search plan with bounded sub-queries and tool-mapped terms.

Updated Apr 15, 2026
One-click install
npx skills add https://github.com/AoManoh/project_templates --skill search-planning
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: search-planning
Source: https://github.com/AoManoh/project_templates/tree/main/skills/search-planning
Command: npx skills add https://github.com/AoManoh/project_templates --skill search-planning

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill prevents research from becoming aimless by turning multi-step investigation into a structured, executable search plan with clear boundaries and traceable outputs.

Core Features & Use Cases

  • Multi-step research planning (6 phases): distill intent, rate complexity, decompose sub-queries, draft search terms, map tools, and set execution order.
  • Boundary guardrails: each sub-query includes a falsifiable boundary and anti-pattern checks to reduce overlap and runaway scope.
  • Evidence archiving: plans for medium-to-large research are archived under docs/references/ with the required phase detail.

Quick Start

Ask your AI agent to research and compare topic A vs topic B systematically, and ensure it first generates a 6-phase search plan and archives the result under docs/references/.

Frequently Asked Questions about search-planning

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I plan a multi-step web search for a comparative analysis?

A structured research plan prevents aimless investigation by decomposing complex questions into bounded sub-queries. It enforces falsifiable boundaries and anti-pattern checks to reduce overlap and runaway scope during multi-step searches.

How do I decompose complex research questions into searchable sub-queries?

Evidence archiving stores your structured research plans under docs/references/. This ensures that medium-to-large comparative or analytical investigations remain fully traceable and on-budget for future verification.

When should I use a structured research plan instead of a simple web search?

Each sub-query in the research plan includes boundary guardrails featuring falsifiable boundaries and anti-pattern checks. These constraints reduce overlap between searches and prevent runaway scope during exploratory investigations.

What is the best way to keep multi-step investigation research traceable?

A structured research planning approach differs from ad-hoc searching by enforcing phase completeness constraints, sub-query falsifiable boundaries, and search term length limits. It maps tools systematically and requires archiving plans into docs/references/.