sciomc

Orchestrate parallel scientist-agent research stages into a verified synthesized report.

Updated Apr 29, 2026
One-click install
npx skills add https://github.com/nichobbs/lyric-lang --skill sciomc-nichobbs
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sciomc
Source: https://github.com/nichobbs/lyric-lang/tree/main/.claude/skills/sciomc
Command: npx skills add https://github.com/nichobbs/lyric-lang --skill sciomc-nichobbs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It coordinates complex research by splitting a goal into multiple stages, running them in parallel, validating the results, and synthesizing a single coherent report.

Core Features & Use Cases

  • Parallel research execution: decomposes goals into 3–7 stages and runs independent “scientist” agents for each stage to gather and analyze evidence efficiently.
  • Verification and conflict detection: cross-validates stage findings to identify contradictions or missing coverage, producing either a verified outcome or conflicts to resolve.
  • AUTO mode autonomy: supports a fully autonomous loop with max-iteration control, state persistence, resume, and completion promises.

Quick Start

Run /oh-my-claudecode:sciomc <your research goal> to automatically decompose the goal, execute staged parallel investigation, verify findings, and synthesize a final report.

Frequently Asked Questions about sciomc

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

FAQPage Schema
How do I run parallel research agents for codebase analysis?

To verify conflicting findings from parallel research agents, a cross-validation loop checks stage results for contradictions or missing coverage, producing either a verified outcome or identified conflicts to resolve before final synthesis.

Can I resume an autonomous research session after it stops?

To verify conflicting findings from parallel research agents, a cross-validation loop checks stage results for contradictions or missing coverage, producing either a verified outcome or identified conflicts to resolve before final synthesis.

What is the best way to decompose a complex codebase security review into stages?

To verify conflicting findings from parallel research agents, a cross-validation loop checks stage results for contradictions or missing coverage, producing either a verified outcome or identified conflicts to resolve before final synthesis.

How does evidence synthesis work when running multi-stage investigations?

To verify conflicting findings from parallel research agents, a cross-validation loop checks stage results for contradictions or missing coverage, producing either a verified outcome or identified conflicts to resolve before final synthesis.

Do I need any external dependencies to orchestrate parallel scientist agents?

To verify conflicting findings from parallel research agents, a cross-validation loop checks stage results for contradictions or missing coverage, producing either a verified outcome or identified conflicts to resolve before final synthesis.

When should I not use autonomous parallel agents for research?

To verify conflicting findings from parallel research agents, a cross-validation loop checks stage results for contradictions or missing coverage, producing either a verified outcome or identified conflicts to resolve before final synthesis.