Deep Research MCP server
{
"mcpServers": {
"fodda-deep-research": {
"type": "http",
"url": "https://mcp.fodda.ai/deep-research"
}
}
}{
"mcpServers": {
"fodda-deep-research": {
"type": "http",
"url": "https://mcp.fodda.ai/deep-research"
}
}
}{
"mcpServers": {
"fodda-deep-research": {
"type": "http",
"url": "https://mcp.fodda.ai/deep-research"
}
}
}{
"servers": {
"fodda-deep-research": {
"type": "http",
"url": "https://mcp.fodda.ai/deep-research"
}
}
}{
"mcpServers": {
"fodda-deep-research": {
"type": "http",
"url": "https://mcp.fodda.ai/deep-research"
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsonversion 1.46.31
What it does
Fodda MCP Server exposes PSFK-curated knowledge graphs covering retail, beauty, sports, fashion, and related domains, along with earnings intelligence and deep research capabilities. It is aimed at developers and AI agents that need structured trend data, brand tracking, and corporate earnings analysis via the Model Context Protocol.
- Run an autonomous deep research session that combines PSFK trend graphs with live web search to produce a citable narrative report on a given topic.
- Query management commentary and analyst Q&A from company earnings calls to detect gaps between executive responses and analyst concerns.
- Track how a specific brand appears across all available knowledge graphs to compile a comprehensive brand intelligence profile.
Tools this server exposes
| Tool | Does |
|---|---|
| list_graphs | Returns metadata and routing instructions for all available knowledge graphs. |
| search_graph | Runs a hybrid keyword and semantic search across curated graphs to surface trend clusters and evidence. |
| search_statistics | Searches for quantitative metrics and numeric data points within the knowledge graphs. |
| search_insights | Retrieves expert quotes, qualitative signals, and professional interpretations from the graphs. |
| get_label_values | Returns valid values for a given node label or category in the graph. |
| get_neighbors | Traverses from seed nodes to discover related concept nodes and their links. |
| get_node | Retrieves detailed metadata and properties for a single node by its ID. |
| get_evidence | Retrieves source articles, citations, and provenance for a trend or node. |
| discover_adjacent_trends | Finds semantically similar trends to a given trend node. |
| get_domain_intelligence | Searches all PSFK curated domain graphs in parallel across retail, beauty, sports, fashion, consumer electronics, and F&B. |
| get_expert_intelligence | Queries specialist industry graphs built by external strategists and experts. |
| get_report_intelligence | Searches institutional report insights from partner organizations such as DHL, PwC, and Delta. |
| get_earnings_intelligence | Queries management commentary, business guidance, and Q&A transcripts from company earnings calls. |
| get_earnings_divergence | Detects deflections and gaps between analyst concerns and executive responses in earnings calls. |
| brand_tracker | Compiles a Brand Intelligence Profile for a given brand across all knowledge graphs. |
| brainstorm_topic | Generates a graph-powered brainstorm map that surfaces unexpected connections and adjacent topic areas. |
| deep_research_topic | Launches an autonomous research session combining knowledge graphs with live web search. |
| check_research_status | Checks the progress of a deep research job or retrieves its final narrative report. |
| get_supplemental_context | Fetches macro context from up to ten institutional data sources in a single query. |
| check_supplemental_status | Retrieves the output from a completed supplemental context job. |
| list_analysts | Lists the available Synthetic Analyst expert personas. |
| consult_analyst | Engages a Synthetic Analyst persona to synthesize a response in a specialized expert voice. |
| generate_visual | Generates branded SVG data visualizations such as Cultural Shifts, Competitive Compass, and Innovation Pathway charts. |
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.