Orchestrate parallel scientist agents to decompose, execute, and synthesize research.
npx skills add https://github.com/dropsyoon/oh-my-claudecode --skill sciomc-dropsyoon
Please help me install this Agent Skill. Skill: sciomc Source: https://github.com/dropsyoon/oh-my-claudecode/tree/main/skills/sciomc Command: npx skills add https://github.com/dropsyoon/oh-my-claudecode --skill sciomc-dropsyoon
sciomc eliminates the time-consuming, step-by-step way of researching complex questions by orchestrating multiple specialist agents in parallel and then verifying and synthesizing their findings into a single structured report.
Give the command /oh-my-claudecode:sciomc AUTO: Analyze the security posture of this authentication system and produce a verified report.
High-intent search queries and answers about installing and using this skill.
Parallel research agents decompose complex architecture analysis goals into 3–7 independent stages, execute them concurrently, and cross-validate findings before synthesizing a final structured report.
Cross-validation during autonomous research workflows detects contradictions, gaps, and missing connections by verifying structured finding, evidence, and confidence tags across parallel agent outputs before final synthesis.
Interrupted parallel agent sessions can be resumed without losing progress because session state is persisted in .omc/research/{session-id}/state.json, enabling status checks and report generation per session-id.
Autonomous execution in AUTO mode supports max-concurrency control to ensure deterministic loop behavior, applying mandatory stage decomposition, state persistence, and explicit completion or blocked promise tags.
Multi-stage decomposition with parallel scientists is needed when sequential research becomes slow or incomplete on complex goals like security analysis, architecture review, and comparative investigation.
Parallel scientist workflows require mandatory stage decomposition and explicit verification promise tags to function correctly, meaning poorly defined research goals or missing evidence tagging will block autonomous execution.
shaun0927Coordinate parallel scientist agents to decompose, execute, verify, and synthesize multi-stage research workflows.
tiankong0101-byteOrchestrates parallel research workflows for complex technical analysis and evidence-backed reporting.
atmigtncaCoordinate parallel scientists to perform autonomous research with decomposition and synthesis.
NovusAevumOrchestrate parallel AI agents to research topics and synthesize reports.
