research-workflow

Structure external research into a mode-based checklist and generate an insights artifact.

11|1|Updated Apr 10, 2026
One-click install
npx skills add https://github.com/Arcanada-one/datarim --skill research-workflow-arcanada-one
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: research-workflow
Source: https://github.com/Arcanada-one/datarim/tree/main/skills/research-workflow
Command: npx skills add https://github.com/Arcanada-one/datarim --skill research-workflow-arcanada-one

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It prevents inconsistent or guess-based planning by structuring external research into a repeatable process that produces decision-ready insights.

Core Features & Use Cases

  • Mode-based research coverage: Runs a full or lite checklist (or skips entirely) depending on task complexity and risk.
  • Tool-aware research and adaptive retrieval: Selects the best available tools (MCP docs, web search/fetch, LTM recall) while always falling back to offline codebase and local documentation analysis.
  • Gap discovery and escalation loop: Detects unknowns during implementation, pauses work, and produces targeted research findings that update the task’s insights.
  • Pre-flight artifact discovery: Reduces planning drift by verifying that named artifacts and assumptions still match the live project state.
  • Provider contract verification: Ensures third-party endpoint input/output shapes are empirically validated before implementation depends on them.

Quick Start

Run the research-workflow skill to generate an insights document for task {task-id} by verifying versions, best practices, breaking changes, compatibility, and any necessary endpoint contracts.

Frequently Asked Questions about research-workflow

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

FAQPage Schema
How do I structure external research to verify tool selection and prevent guess-based planning?

Structuring external research into a mode-based checklist prevents guess-based planning by verifying tool selection, checking versions, and documenting decision-ready insights into a repeatable artifact.

How do I validate third-party endpoint contracts before implementation depends on them?

Validating third-party endpoint contracts requires provider verification probes that empirically test input and output shapes, ensuring endpoint compatibility before implementation depends on them.

What is the best way to discover knowledge gaps during requirements analysis?

Discovering knowledge gaps during requirements analysis uses an escalation loop that pauses work upon detecting unknowns, producing targeted research findings that update the task's documented insights.

Can I run a lite research checklist for low complexity tasks instead of a full scan?

You can run a lite research checklist or skip it entirely for low complexity tasks, as complexity-aware routing adjusts the research coverage and tool selection based on project risk.

How do I verify that named artifacts and assumptions still match the live project state?

Verifying named artifacts and assumptions against the live project state is achieved through pre-flight artifact discovery, which reduces planning drift before implementation begins.

What should I do when web search and MCP docs are unavailable during tool selection?

When web search and MCP docs are unavailable, tool selection adapts by falling back to offline codebase analysis and local documentation checks to complete the research workflow.