From geepers-mcp
Launch parallel multi-domain search workflows using the Dream Swarm orchestrator. Use when (1) searching across multiple data sources simultaneously, (2) aggregating information from APIs, databases, and web sources, (3) comparing findings across domains, (4) rapid parallel data gathering. Requires the MCP server to be running.
npx claudepluginhub lukeslp/geepers-mcp --plugin geepers-mcpThis skill uses the workspace's default tool permissions.
Launch parallel multi-agent search workflows across multiple domains.
Guides agents through structured research phases: planning queries, multi-query execution, source analysis, and synthesis for comprehensive investigations or reports.
Conducts deep web research with parallel agents, multi-wave exploration for gaps, and structured synthesis. Activates for investigating topics, comparing options, best practices, or comprehensive web info.
Conducts systematic internet research via strategic questioning, multi-source analysis, credibility evaluation with tier system, and structured .research/ reports. Use for best practices, comparisons, technology evaluations, trends.
Share bugs, ideas, or general feedback.
Launch parallel multi-agent search workflows across multiple domains.
┌────────────────────────┐
│ SEARCH QUERY │
│ "AI safety research" │
└───────────┬────────────┘
│
┌─────────────────┼─────────────────┐
│ │ │
┌────▼────┐ ┌────▼────┐ ┌────▼────┐
│ arXiv │ │ GitHub │ │ News │
│ Agent │ │ Agent │ │ Agent │
└────┬────┘ └────┬────┘ └────┬────┘
│ │ │
└─────────────────┼─────────────────┘
│
┌───────────▼───────────┐
│ AGGREGATOR │
│ Dedupe & Synthesize │
└───────────────────────┘
| Domain | Description | Example Query |
|---|---|---|
arxiv | Academic papers | "quantum computing" |
github | Code repositories | "react components" |
news | Recent articles | "AI regulation 2024" |
wikipedia | Encyclopedia | "machine learning" |
pubmed | Medical literature | "CRISPR therapy" |
semantic_scholar | Academic index | "transformer architectures" |
census | US demographics | "population density" |
nasa | Space/Earth data | "mars rover" |
youtube | Video content | "tutorial python" |
weather | Forecasts | "Seattle forecast" |
openlibrary | Book metadata | "science fiction" |
fec | Campaign finance | "2024 donations" |
judiciary | Court records | "antitrust case" |
archive | Wayback Machine | "historical websites" |
finance | Market data | "AAPL stock" |
mal | Anime database | "studio ghibli" |
wolfram | Computation | "integral sin(x)" |
scripts/swarm-search.py "AI safety research" --domains arxiv github news
scripts/swarm-search.py "climate change" --agents 10 --parallel 5
scripts/swarm-search.py "quantum computing" --all-domains
scripts/swarm-status.py workflow-xyz789
scripts/swarm-status.py workflow-xyz789 --results
| Parameter | Default | Description |
|---|---|---|
--domains | arxiv,news | Comma-separated domain list |
--agents | 5 | Number of parallel agents (1-20) |
--parallel | 3 | Max concurrent domain searches (1-10) |
--provider | xai | LLM provider for synthesis |
--max-results | 10 | Results per domain |
--stream | false | Enable SSE streaming |
--output | stdout | Output format (stdout, json, markdown) |
scripts/swarm-search.py "WebAssembly optimization" \
--domains arxiv,github,semantic_scholar \
--agents 6 --parallel 3
scripts/swarm-search.py "AI regulation policy" \
--domains news,wikipedia,youtube \
--agents 5 --output markdown
scripts/swarm-search.py "CRISPR gene therapy" \
--domains arxiv,pubmed,news,wikipedia \
--agents 8 --parallel 4 --stream
{
"task_id": "swarm-xyz789",
"status": "completed",
"results": {
"arxiv": [
{"title": "...", "url": "...", "abstract": "..."}
],
"github": [
{"name": "...", "url": "...", "description": "..."}
],
"news": [
{"title": "...", "url": "...", "source": "..."}
]
},
"synthesis": {
"summary": "...",
"key_themes": [...],
"cross_domain_insights": [...]
},
"metadata": {
"domains_searched": 3,
"total_results": 24,
"execution_time": 12.5
}
}
| Aspect | Dream Swarm | Dream Cascade |
|---|---|---|
| Speed | Faster (parallel) | Slower (hierarchical) |
| Depth | Broad coverage | Deep analysis |
| Use case | Data gathering | Research synthesis |
| Agent count | Many simple agents | Fewer complex agents |
| Cost | Lower | Higher |
"Domain not available": Check API key configured for that domain
"Rate limited": Reduce --parallel count
"No results": Try broader query or different domains