{
  "markdown": "# Financial Signals MCP\n\n**Derived financial intelligence for AI agents — not another SEC EDGAR filings\nserver.** Six of those already exist. This serves the *interpreted* layer: the\npatterns, surprises, flows, scores, and anomalies an agent actually needs to make\na decision. Raw data is commodity; **interpreted signals are premium**.\n\n> Part of the **FoundryNet Data Network**. Every output carries verifiable\n> provenance. See also:\n> **gov-contracts-mcp**, **brand-intel-mcp**, **patent-intel-mcp**.\n>\n> Built as a **free-tier alternative to enterprise financial data** (FactSet,\n> Morningstar, S&P Capital IQ) — financial intelligence for agents without\n> enterprise subscriptions.\n\n## Connect\n\n- **MCP endpoint** (Streamable HTTP): `https://financial-signals-mcp-production.up.railway.app/mcp`\n- **Registry**: `io.github.FoundryNet/financial-signals-mcp`\n- **Agent card**: `https://financial-signals-mcp-production.up.railway.app/.well-known/agent-card.json`\n\n### Claude Desktop / Cursor / Claude Code\n\n```\nclaude mcp add --transport http financial-signals https://financial-signals-mcp-production.up.railway.app/mcp\n```\n\n```json\n{ \"mcpServers\": { \"financial-signals\": { \"url\": \"https://financial-signals-mcp-production.up.railway.app/mcp\" } } }\n```\n\n## Tools\n\n| Tool | Price | What it does |\n|---|---|---|\n| `insider_activity` | $0.01 | Insider transactions **with pattern analysis** (cluster_sell, large_buy, ceo_buy, pre_earnings) |\n| `earnings_check` | $0.01 | 8-quarter EPS surprises, beat streak, guidance trend, next date |\n| `institutional_moves` | $0.01 | Significant 13F position changes with context |\n| `screen_stocks` | $0.01 | Ratio screen ranked by the **proprietary composite_value_score** |\n| `sector_snapshot` | $0.01 | Sector medians, top/bottom by value score, aggregate trend |\n| `macro_dashboard` | **free** | Macro indicators with trend + historical percentile — the gateway |\n| `company_profile` | $0.01 | Full blended profile + value score + sector positioning |\n| `anomaly_alert` | $0.02 | Unusual patterns across all monitored companies (premium) |\n| `daily_brief` | $25 | Curated daily brief: top insider/earnings/institutional/value-score movers + macro |\n| `brief_summary` | $0.50 | Top 5 signals from today's brief as structured JSON (cheap sample of `daily_brief`) |\n| `mint_info` | **free** | FoundryNet Data Network + provenance details |\n\n**Free tier:** 25 paid-tool queries/day per agent (plus free `macro_dashboard` +\n`mint_info`). Then the tool returns an HTTP-402 with a metered payment challenge —\npay it, re-call with the same args plus `payment_tx=<reference>`. An\n`Authorization: Bearer fnet_…` key bypasses the paywall.\n\n## The moat: composite_value_score\n\n`screen_stocks` and `company_profile` expose a proprietary **composite_value_score\n(0-100)** — a transparent blend of: valuation vs. sector, growth, margin quality,\ninsider sentiment, institutional momentum, and earnings consistency. Sort\n`screen_stocks` by it and you're using a ranking nobody else computes.\n\n## Sources & method\n\nA daily run after US market close computes DERIVED tables (not raw filings) over\nthe S&P 500 from free sources: **yfinance** (price, ratios, earnings, insider &\ninstitutional — Yahoo's aggregation of SEC data), **FRED** (macro), and **SEC\nEDGAR** (supplementary). Signal-typing + the composite score are computed at ingest.\n\n**Honesty notes:** revenue surprise, guidance direction, and post-earnings move\nare not reliably free, so they're null/`none` where unavailable (`roic` uses ROA\nas a capital-efficiency proxy). Coverage grows as the daily universe is processed.\n\n## More\n\npart of the **FoundryNet Data Network**.\n\nBuilt by [FoundryNet](https://foundrynet.io?utm_source=github&utm_medium=readme&utm_campaign=financial-signals-mcp) · forge@foundrynet.io\n\n## Live network activity\n\n**Live feed:** [mint.foundrynet.io/feed](https://mint.foundrynet.io/feed)  \nReal-time verified work across 17 servers and autonomous agents, with cryptographic provenance.\n",
  "bytes": 3998,
  "sha": "16e50844dd32e9109c50027bc4c664ce990199ab18e56f76e717d925a552c188",
  "repo_slug": "foundrynet/financial-signals-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_foundrynet_financial_signals_m_9206b253/readme"
}