By basher83
Multi-agent research pipeline with parallel researchers and synthesis
Find official documentation and project architecture
Semantic search for conceptually related content
Find repositories, implementations, and code patterns
Web research via Jina parallel search
Combines findings from all researcher reports into unified synthesis
External network access
Connects to servers outside your machine
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Multi-agent research pipeline for Claude Code. Orchestrates 5 specialized researcher agents (GitHub, Tavily, DeepWiki, Exa, Jina) and synthesizes findings with source authority hierarchy.

Add the lunar-claude marketplace, then install the plugin:
/plugin marketplace add basher83/lunar-claude
/plugin install lunar-research@lunar-claude
Set your API keys as environment variables. A SessionStart hook substitutes them into the MCP configuration automatically.
Add to ~/.bashrc, ~/.zshrc, or your shell's config:
export GITHUB_TOKEN="ghp_xxxxxxxxxxxx"
export TAVILY_API_KEY="tvly-xxxxxxxxxxxx"
export EXA_API_KEY="xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
export JINA_API_KEY="jina_xxxxxxxxxxxx"
Run source ~/.bashrc or restart your terminal.
Create .mise.local.toml in your project directory:
[env]
GITHUB_TOKEN = "ghp_xxxxxxxxxxxx"
TAVILY_API_KEY = "tvly-xxxxxxxxxxxx"
EXA_API_KEY = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
JINA_API_KEY = "jina_xxxxxxxxxxxx"
Run mise trust to activate. The .local.toml suffix is gitignored.
/lunar-research:run "your research query"
/lunar-research:cache # List cached research
/lunar-research:cache --detailed # Show researcher status
| Command | Description |
|---|---|
/lunar-research:run | Main research orchestrator |
/lunar-research:cache | List cached research entries |
| Agent | Purpose | MCP Tools |
|---|---|---|
github-researcher | Find repositories and code patterns | GitHub MCP |
tavily-researcher | Find tutorials and community content | Tavily MCP |
deepwiki-researcher | Find official documentation | DeepWiki MCP |
exa-researcher | Semantic search for related content | Exa MCP |
jina-researcher | Search web and arXiv for academic content | Jina MCP |
synthesizer-agent | Combine findings into synthesis | None |
research-report.schema.json - JSON Schema for researcher output formatvalidate-research-report.py - Validate reports against schemasetup-mcp-keys.sh - API key configuration (if hook fails)SessionStart - Substitutes API keys from environment variables into .mcp.json on each
session start. Run /mcp to reconnect if keys change mid-session.The plugin caches research results and shares them across projects:
${CLAUDE_PLUGIN_ROOT}/cache/
├── index.json # Knowledge base index (tracked in git)
└── [normalized-query]/ # Per-query cache (gitignored)
├── github-report.json
├── tavily-report.json
├── deepwiki-report.json
├── exa-report.json
├── jina-report.json
└── synthesis.md
The plugin provides five MCP servers (configured by the SessionStart hook):
| Server | Tools Prefix | Auth |
|---|---|---|
| GitHub | mcp__plugin_lunar-research_github__* | GITHUB_TOKEN |
| Tavily | mcp__plugin_lunar-research_tavily__* | TAVILY_API_KEY |
| DeepWiki | mcp__plugin_lunar-research_deepwiki__* | None |
| Exa | mcp__plugin_lunar-research_exa__* | EXA_API_KEY |
| Jina | mcp__plugin_lunar-research_jina__* | JINA_API_KEY |
See MIT
npx claudepluginhub basher83/lunar-claude --plugin lunar-researchEasily create hooks to prevent unwanted behaviors by analyzing conversation patterns
Comprehensive plugin development toolkit for creating Claude Code plugins with AI-powered guidance for skills, agents, commands, and hooks
Human-in-the-loop Architecture Decision Record workflow
Git workflow automation with fork-isolated skills for atomic commits, branch cleanup, conventional commit standards, and context-aware PR review
End-to-end Ansible workflow automation: design, create, validate, and review playbooks and roles with orchestrated multi-agent pipelines
Deep research coordination: academic papers, technical analysis, data insights, and web intelligence
Multi-source research plugin — code archaeology, community discourse, academic literature, and TRIZ cross-domain analysis with domain-adaptive depth
Multi-agent deep research plugin with parallel web searches and synthesis
Autonomous, personalized research loops for Claude Code. Set a topic, walk away, come back to a quality-gated report adapted to your projects.
Agent skill and CLI helpers for using RivalSearchMCP research tools from Claude Code. Includes workflows, command references, and a standalone CLI for the hosted MCP server.
AI-powered deep research with multi-agent source verification and structured outputs