Skip to content

160-tool MCP server for real-time global intelligence from free public APIs: markets, liquidity, prediction markets, SEC filings, conflict, military, aircraft-cohort early warning, migration and border flows, cyber, climate, news and more. Dated observation history, 29 research prompts, cited briefs, geofences, SSE dashboard. MIT, no paid API keys.

Merge feat/news-market-reaction: news vs market reaction (Phase 29)

Phoenix Agent (mac-studio)008d97dd625c
diff --cc CHANGELOG.md
index 629a2fa,f675f0c..5836d95
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@@ -2,60 -2,32 +2,86 @@@
## Unreleased
+### Added: liquidity and money (Phase 29)
+Keyless sources only: the FRED graph CSV (`fred.stlouisfed.org/graph/fredgraph.csv`,
+allowed by robots.txt with `Crawl-delay: 1`, honored by a 1 s `fred-graph` rate
+floor) and the ECB Data Portal (SDMX-JSON). Every point carries `observed_at`,
+`value` and `units`; every series is typed `amount_outstanding`, `flow`,
+`growth_rate` or `index` and carries a freshness block that flags data older than
+its publication cadence allows (a served stale cache copy or an outage shows up
+there). `published_at` is null: neither payload carries release dates, and the
+responses say so. `limit=0` returns the full history.
+- `intel_net_liquidity`: weekly Fed total assets (WALCL) minus Treasury General
+ Account (WDTGAL) minus overnight reverse repo (RRPONTSYD), in USD billions.
+ WALCL and WDTGAL are USD millions and are divided by 1,000; RRPONTSYD is already
+ USD billions. One point per WALCL Wednesday; the TGA must be reported for that
+ Wednesday; RRP comes from that day or, when that day is blank (a holiday), the
+ latest earlier business day within 5 days, and each point names the RRP date
+ used. Weeks without aligned components are skipped and counted, never filled
+ with zero, and a failed component makes the whole response an error. The TGA
+ term is the Wednesday level (WDTGAL), not the week average (WTREGEN). A
+ `method` block, returned with errors too, states each term's observation basis
+ (WALCL and WDTGAL Wednesday levels in USD millions, RRPONTSYD daily in USD
+ billions), its conversion, the RRP alignment and why WTREGEN is not used.
+- `intel_liquidity_board`: latest reading and change vs the previous observation
+ for the Fed balance sheet, TGA, RRP, net liquidity, US M2, euro area M1/M2/M3
+ stocks and Eurosystem total assets, sortable by `biggest_move` or `newest`. A
+ failed source becomes an `unavailable` row; net liquidity is dropped, not
+ computed with a zero term, when any of its components fails. A `blocked` list
+ names each measure not covered with its publisher and the status measured on
+ 2026-10-17.
+- `intel_money_supply`: US M2 (FRED M2SL) and euro area M1, M2 and M3 from the ECB
+ BSI dataset (end-of-month stocks, monthly flows, annual growth rates), kept
+ separate and never summed. US M3 is reported as discontinued (FRED M3SL ends at
+ 2006-02-01).
+- `intel_fed_liquidity_series`: WALCL, WDTGAL, WTREGEN, RRPONTSYD or M2SL with full
+ history; blank source dates (2,833 for RRPONTSYD on 2026-10-17, 2,679 of them
+ before 2013-09-23, when daily values begin) are counted, not read as zero.
+ Net liquidity therefore starts at 2003-02-12 with 523 earlier or unaligned
+ weeks skipped and counted, the last of them 2013-09-18.
+- `intel_ecb_series`: any ECB BSI or ILM series by SDMX key, with units, scale,
+ titles and provisional flags taken from the response. A growth rate's change is
+ reported in points, not as a percent of a percent.
+- CLI: `intel liquidity`, `intel net-liquidity`, `intel money-supply`,
+ `intel fed-liquidity`, `intel ecb-series`. Collector: `liquidity_board` and
+ `net_liquidity` (group `liquidity`).
+- Live reading 2026-10-17: net liquidity for Wednesday 2026-10-14 was
+ USD 5,869.615 billion (WALCL 6,755,639 million, WDTGAL 885,823 million,
+ RRPONTSYD 0.201 billion, all observed 2026-10-14).
+
+Not added, measured 2026-10-17: ICI money market fund assets and ETF net issuance
+(403 on every path, robots.txt included); FINRA margin statistics (the finra.org
+Terms of Use, Restrictions d and e, forbid building a database from the site and
+data mining, scraping or harvesting tools, robots included); BLS public API (keyless
+v1 answers for CPI-U and payrolls, but api.bls.gov robots.txt disallows all paths
+and the terms page answers 403). Fund flows and ETF net issuance therefore have no
+source here, and no proxy is substituted for any of these measures.
+ ### Added: news vs market reaction
+ - `intel_news_market_reaction` (CLI: `intel news-reaction`): the companies and
+ countries named in recent story clusters, mapped to a listing (the entity
+ table's ticker, confirmed against SEC's company_tickers.json for US
+ listings) or to the country's primary index (the table `intel_country_stocks`
+ uses), with the move on the story's market session from Yahoo daily bars.
+ Company rows give the stock's close-to-close return (adjusted closes), the
+ benchmark's return over the same two dates (SPY for US listings, the
+ exchange country's index otherwise) and the excess return; country rows
+ give the index return and a z-score against its previous up-to-20 sessions.
+ - Session rule, stated in every response: the story time (earliest
+ `published` in the cluster) is converted to the exchange's timezone; before
+ the regular close on a trading day it is that session, at or after the close
+ the next one, and a weekend or holiday (read from the listing's own bars)
+ maps to the next trading day. A session that has not closed is `pending`.
+ - Honest gaps: a failed quote, a missing benchmark bar, a stock series missing
+ a day its benchmark traded, a story timestamp later than the collection time
+ and a story older than the bar range are each a `data_gap` row with the
+ reason and no number, never a 0% move. Unmapped names (private companies,
+ countries without an index) are listed, not dropped. Company names that are
+ also common words (apple, amazon, meta, shell, bp, alphabet, chevron,
+ samsung) are excluded with a stated reason unless the headlines carry a
+ corroborating term or an article comes from a feed category where the
+ company sense dominates (a hyphen-joined use such as "meta-analysis" does
+ not count). Responses say they describe same-session co-movement, not
+ causation, and are not a trading signal. Tools 139 -> 140.
## 0.11.0 - 2026-10-12
diff --cc README.md
index b88d20f,e285433..a54efc1
--- a/README.md
+++ b/README.md
@@@ -6,7 -6,7 +6,7 @@@
[![Python 3.11+](https://img.shields.io/badge/Python-3.11%2B-green)](https://python.org)
[![License](https://img.shields.io/badge/License-MIT-yellow)](LICENSE)
- Real-time global intelligence across **30+ domains** with **144 MCP tools**, a live ops-center dashboard, a CLI, and a **Qdrant vector store** for enterprise-grade semantic search across accumulated intelligence. All data comes from free, public APIs: no paid subscriptions required.
-Real-time global intelligence across **30+ domains** with **140 MCP tools**, a live ops-center dashboard, a CLI, and a **Qdrant vector store** for enterprise-grade semantic search across accumulated intelligence. All data comes from free, public APIs: no paid subscriptions required.
++Real-time global intelligence across **30+ domains** with **145 MCP tools**, a live ops-center dashboard, a CLI, and a **Qdrant vector store** for enterprise-grade semantic search across accumulated intelligence. All data comes from free, public APIs: no paid subscriptions required.
Built for AI agents that need world awareness: market conditions, geopolitical risk, military posture, aircraft-cohort early warning, refugee and migration flows, supply chain disruptions, cyber threats, and more — all queryable via the Model Context Protocol. The vector store enables natural language queries like *"military activity near Taiwan"* or *"cyber threats targeting healthcare"* across all historical data.
@@@ -65,7 -65,7 +66,7 @@@
| **Tropical Cyclones** | 2 | NHC (Atlantic, E/C Pacific) + JTWC (NW Pacific, N Indian, S Hemisphere) |
| **Situation Brief** | 1 | Cited situational awareness brief over MCP: bounded server-side overview synthesized via local Ollama, with a mechanically-cited fallback |
- **Total: 144 tools** across 30+ intelligence domains.
-**Total: 140 tools** across 30+ intelligence domains.
++**Total: 145 tools** across 30+ intelligence domains.
---
@@@ -130,8 -130,7 +131,9 @@@ intel markets # stock indi
intel earthquakes --min-mag 5.0
intel migration-surges # year-over-year displacement jumps (UNHCR)
intel migration-flows -o Sudan
+intel liquidity # Fed balance sheet, TGA, RRP, net liquidity, US M2, euro area M1-M3
+intel net-liquidity -n 0 # full weekly history, components and RRP date per week
+ intel news-reaction -c finance # same-session moves of companies in the news
intel status # cache + circuit breaker health
```
diff --cc ROADMAP.md
index 6a4cbfd,43cca17..7ca117d
--- a/ROADMAP.md
+++ b/ROADMAP.md
@@@ -577,9 -577,9 +577,9 @@@ dashboard
| 1 | Research guide over MCP | MCP resources (guide generated from the live registry and source health, measures glossary, "what this server will not do") and MCP prompts (worked research questions, each naming the tools it needs and the limit to respect). Server exposed no prompts or resources before | :red_circle: |
| 1 | Hostility timeline | Per country pair or country, daily counts of threat / sanction / force / criticism coverage plus tone; sortable by most covered, most hostile, biggest change | :red_circle: |
| 1 | Story breadth | News clusters carry a distinct-outlet count; a top-stories view ranks by how many outlets carry a story | :red_circle: |
-| 1 | Liquidity and flows | Fed balance sheet, Treasury General Account, reverse repo and net liquidity (FRED keyless CSV), money supply (US M2, euro area M1/M2/M3, never summed), ECB, ICI fund flows and ETF net issuance, FINRA margin statistics, BLS series; amounts kept separate from flows | :red_circle: |
+| 1 | Liquidity and flows | Fed balance sheet, Treasury General Account, reverse repo and net liquidity (FRED keyless CSV), money supply (US M2, euro area M1/M2/M3, never summed), ECB, ICI fund flows and ETF net issuance, FINRA margin statistics, BLS series; amounts kept separate from flows | :yellow_circle: Built, unreleased: 5 tools (`intel_liquidity_board`, `intel_net_liquidity`, `intel_money_supply`, `intel_fed_liquidity_series`, `intel_ecb_series`), 5 CLI commands, 2 collector sources. FRED graph CSV and ECB Data Portal live-verified 2026-10-17. The TGA term is WDTGAL (Wednesday level, USD millions): WTREGEN is a week average; the net liquidity `method` block states each term's units, conversion and how RRPONTSYD (daily, USD billions) is aligned to the Wednesday. Blocked, measured 2026-10-17, and listed in the board's `blocked` field; no proxy is substituted, so fund flows and ETF net issuance have no source: ICI (money market fund assets at /research/stats/mmf, ETF net issuance at /research/stats/etf) answers 403 Akamai "Access Denied" on every path, robots.txt included; FINRA margin statistics XLSX answers 200, but the finra.org Terms of Use (Restrictions d and e) forbid building a database from the site and data mining, scraping or harvesting tools, robots included, and the page says data feeds are not available; BLS public API v1 answers 200 without a key (CPI-U, payrolls), but api.bls.gov robots.txt is `Disallow: /` and the terms page on www.bls.gov answers 403 to this server's User-Agent, so it is not used |
| 1 | Kalshi + cross-venue markets | Kalshi public market data next to Polymarket; same-event comparison with an explicit match confidence; settled markets scored against pre-settlement prices (Brier) | :red_circle: |
- | 1 | News vs market reaction | Companies and countries named in story clusters mapped to tickers / country indices; same-day move against a benchmark | :red_circle: |
+ | 1 | News vs market reaction | Companies and countries named in story clusters mapped to tickers / country indices; same-day move against a benchmark | :white_check_mark: Built 2026-10-17, unreleased: `intel_news_market_reaction` and `intel news-reaction`. Company rows: stock, benchmark (SPY, or the exchange country's index) and excess return on the story's session; country rows: index return with a trailing z-score. Session rule in the exchange's timezone (after close = next session; weekends and holidays read from the listing's own bars). Open: `^TASI` and `^TA125` (Saudi and Israel in the country-index table) return 404 on Yahoo, so those benchmarks are data gaps; story times inherit `sources/news.py`'s RSS timestamp offset until that is fixed |
| 2 | Connections with timing offsets | Correlation between retained series over a shared sample, tested at leads and lags, with minimum sample size and false-discovery control; states that it does not prove causation | :red_circle: |
| 2 | Day timeline | What every retained source recorded for a given day, by observation date or collection date | :red_circle: |
| 3 | Dashboard views | Liquidity board, situations, top stories, collection health panels | :red_circle: |
diff --cc src/world_intel_mcp/tools/__init__.py
index 29be168,ea58643..be4a2a3
--- a/src/world_intel_mcp/tools/__init__.py
+++ b/src/world_intel_mcp/tools/__init__.py
@@@ -30,8 -30,8 +30,9 @@@ from . import ( # noqa: E40
hazards,
infrastructure,
intelligence,
+ liquidity,
markets,
+ news_market,
society,
synthesis,
system,
@@@ -50,7 -50,7 +51,8 @@@ _MODULES =
geospatial,
synthesis,
finance,
+ liquidity,
+ news_market,
vector,
aoi,
system,