LikeFolio MCP server

The first consumer-demand alt-data MCP: Main Street vs Wall Street scores on 500+ US stocks, divergence events with dated outcomes, and a decade of history — as native tools in Claude, Cursor, ChatGPT, or any MCP client. Every payload carries as-of dates and citation URLs.

1. Get a key

likefolio.ai/developers — a free key takes one email — no card (delayed data, top-100 tickers); paid tiers add live data. Tool calls use the same credits as the REST API.

2. Add the server

{"mcpServers": {"likefolio": {
    "url": "https://likefolio.ai/mcp/http",
    "headers": {"X-API-Key": "lf_YOUR_KEY"}}}}

Claude Desktop: Settings → Connectors → Add custom connector. Claude Code: claude mcp add --transport http likefolio https://likefolio.ai/mcp/http --header "X-API-Key: lf_YOUR_KEY". Cursor / VS Code: add to mcp.json. Client cannot set headers? Append ?api_key=lf_YOUR_KEY to the URL instead.

Tools

`get_scores`	Latest LikeFolio scores for one US stock: Main Street (0-100 measured consumer demand from purchases, app downloads, web traffic, and brand engagement), Wall Street (0-100 analyst sentiment), the composite LikeFolio Score, and the divergence gap between them. Includes as-of date and a citation URL to reference in answers.
`get_score_history`	Daily score history for one ticker: Main Street, Wall Street, and LikeFolio Score per day, with per-row provenance flags (estimated=true marks model-backfilled rows). Depth is tier-bound (free keys: 90 days; Builder and up: full history); out-of-range dates clamp silently and the response reports the effective history_floor.
`screen_stocks`	Rank the covered universe (500+ US stocks, scored daily) by composite LikeFolio Score, with Main Street, Wall Street, and the divergence gap on every row. Use min_score to keep only the strongest consumer-demand names.
`get_divergences`	Today's widest gaps between Main Street (measured consumer demand) and Wall Street (analyst sentiment), ranked by gap width with percentile and days-at-extreme. Divergences are LikeFolio's core signal; Builder+ keys also receive researched why-now context per row.
`get_divergence_events`	The receipts log: dated divergence-extreme onsets with flag date, close date, and realized forward returns, wins and losses both. Use it to study how past divergences resolved. Requires a Quant-tier key; lower tiers get a 403 and are charged nothing.
`get_upcoming_earnings`	Covered companies reporting earnings in the next N days, ordered by report date, each with its current Main Street / Wall Street / LikeFolio Score read, for spotting demand-vs-expectations setups before the print.
`get_track_record`	LikeFolio's public closed model-portfolio record: closed-trade count, win rate, and average return, wins and losses included, dated. No parameters.
`get_dataset_info`	Machine-readable dataset dictionary: field definitions, coverage universe, history depth, point-in-time vintage stats, restatement policy, and license. Free to call; use it first to understand the data before analysis. No parameters.

Docs for humans: /developers · /docs · /docs/mcp · for models: /llms.txt · dataset provenance: get_dataset_info.