Hosungyou Diverga
hosungyou-diverga is a Claude Code plugin for the Idea phase that delivers 44 specialized research agents with VS methodology, human checkpoints, creative alternatives, and parallel execution.
by HosungYou · github.com/HosungYou/Diverga
Orchestrate 44 specialized research agents with VS methodology, human checkpoints, creative alternatives, and parallel execution before committing to a build.
Add it to Claude Code
Install the plugin in Claude Code. One command, paste-ready.
/plugin install hosungyou-diverga@HosungYou/DivergaBuilt to be called by your agent
Skillselion is itself an MCP server. Your agent can pull this entry and a paste-ready install config straight from the API - no copy-paste.
Retrieve this entry with skillselion.get_details("plugin:HosungYou/Diverga") and the paste-ready config with skillselion.get_install_config("plugin:HosungYou/Diverga").
What it does
hosungyou-diverga is a Claude Code plugin packaging Diverga: forty-four specialized AI research agents governed by VS methodology that emphasizes human-centered checkpoints, creative alternatives, and parallel execution. Solo builders and analysts install it when early journey work needs structured, multi-agent research rather than a single generic web search prompt—competitive scans, literature-style synthesis, meta-analysis framing, and scoped decisions before prototype spend. Canonical placement is idea research because the catalog positions a complete research agent system across categories; validate scope and operate iterate are natural follow-ons when checkpoints narrow what to build or when you re-run research after shipping. The workflow is methodology-heavy: expect explicit human gates and branching alternatives, not unattended autopilot. Complexity is advanced due to orchestrating many agents and interpreting VS outputs. After registration, you gain a coordinated research stack inside Claude Code instead of hand-picking disparate research skills one topic at a time.
Highlights
- Suite of 44 specialized AI research agents organized across research categories
- VS methodology with human-centered checkpoints before conclusions harden
- Creative alternatives generation so agents do not converge on one shallow answer
- Parallel execution for faster multi-agent literature and evidence passes
- Single-plugin Diverga bundle described as a complete research agent system
Why builders use it
Shallow one-shot AI research misses nuance, skips human judgment, and cannot run many specialized agents in parallel when you are still deciding what to build.
After installing Diverga, you can run categorized research agents under VS methodology with checkpoints and parallel jobs, producing alternatives-backed evidence for scope and build decisions.
At a glance
- Type - Plugin in AI Agents.
- Adoption - 0 installs, 2 stars, 0 votes.
FAQ
Who is hosungyou-diverga for?
It is for solo builders, analysts, and small teams who need deep, multi-agent research with human checkpoints and creative alternatives during discovery and scoping.
When should I use hosungyou-diverga?
Use it in the idea and validate phases when exploring markets, synthesizing evidence, or running parallel specialized agents before you commit engineering time.
How do I add hosungyou-diverga to my agent?
Register the HosungYou/Diverga Claude Code plugin, then invoke the Diverga research agent suite and follow VS methodology checkpoints inside your Claude Code session.
Comments
Share how you use hosungyou-diverga, gotchas, or tips for other indie builders.
No comments yet - be the first to share how you use it.