research

Decompose research goals into parallel stages and synthesize validated reports.

Updated Feb 20, 2026
One-click install
npx skills add https://github.com/ssccio/cc-on-web-base --skill research-ssccio
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: research
Source: https://github.com/ssccio/cc-on-web-base/tree/main/skills/research
Command: npx skills add https://github.com/ssccio/cc-on-web-base --skill research-ssccio

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Research goals often take too long to investigate deeply and consistently, especially when you need broad coverage, cross-validation, and a synthesized report.

Core Features & Use Cases

  • Parallel scientist workflow: Decomposes a research goal into multiple independent stages and runs them concurrently to speed up discovery.
  • Verification loop: Cross-validates stage outputs to detect contradictions, coverage gaps, and missing connections before reporting.
  • AUTO mode session orchestration: Enables autonomous iterative execution with explicit promise tags, session state persistence, resume, and cancellation.
  • Evidence- and confidence-structured outputs: Encourages reproducible findings using tagged evidence blocks, confidence levels, and cited file context.
  • Report generation: Produces a consolidated research report template including methodology, key findings, limitations, and recommendations.

Quick Start

Run: /oh-my-claudecode:research AUTO: Analyze authentication patterns in this codebase.

Frequently Asked Questions about research

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

FAQPage Schema
How do I run parallel research agents to speed up comprehensive report generation?

Parallel research agents accelerate comprehensive report generation by decomposing a single research goal into independent stages that execute concurrently. This parallel scientist workflow speeds up discovery and cross-validates outputs to produce a structured, consolidated report.

What is cross-validation in parallel research and how does it verify findings?

Cross-validation in parallel research works by running a verification loop that checks concurrent stage outputs against each other. It detects contradictions, identifies coverage gaps, and finds missing connections before synthesizing the final report, ensuring findings are consistent and reliable.

How do I orchestrate autonomous iterative research sessions with session control?

Autonomous iterative research sessions use AUTO mode session orchestration with explicit promise tags to manage execution. This mechanism provides session state persistence, allowing you to resume, iterate, or cancel ongoing parallel research workflows reliably across stages.

Can I structure research outputs with evidence and confidence levels for reproducible findings?

Yes, structured research outputs use tagged evidence blocks, confidence levels, and cited file context to encourage reproducible findings. This evidence-confidence formatting ensures that every synthesized finding in the final report template is traceable and verifiable.

What are the limitations of using automated agent orchestration for research decomposition?

Automated agent orchestration for research decomposition requires YAML-frontmatter-discovered invocation and explicit completion or blocking semantics. Without proper session state persistence and promise tags, autonomous iterative execution may fail to reliably manage complex, concurrent research stages.