HillSignal

Federal contracts, SAM notices, congressional trades, PAC money, insider trades, bills, EOs, themes.

我該用這個嗎

品質與安全性

A
說明品質
94%
結構描述完整度
85%
命名品質
87%
汙染風險
100%
權限相符程度
100%
協定合規性
100%

發現項目(3)

  • LOWTool 'resolve_entity' description lacks action verb在 resolve_entity 中
  • LOWTool 'get_politician' description lacks action verb在 get_politician 中
  • LOWTool 'list_themes' description lacks action verb在 list_themes 中

根據工具定義與協定合規性的自動化分析。

上下文成本

~4,467Token(工具定義)
~1.1 KB典型回應大小
顯著的注意力影響(128k 上下文的 3.49%)

這是每次將伺服器的工具載入模型上下文時所消耗的約略 token 數量。數量越高,可用於其他工作的注意力就越少。

安裝

一鍵安裝

將以下內容加入你的 `claude_desktop_config.json` 檔案:

{
  "mcpServers": {
    "hillsignalapp": {
      "url": "https://www.hillsignal.com/api/mcp"
    }
  }
}

遠端端點

https://www.hillsignal.com/api/mcpstreamable-http

它能做什麼

工具清單

工具(24)

🟢 唯讀🟡 寫入🔴 刪除⚪ 未知
🟢resolve_entity(name)

Map an organization name (a contract recipient, a subsidiary, a PAC sponsor, a Form D issuer) to its public-company ticker. Exact normalized match against SEC legal names, aliases and the Exhibit 21 subsidiary map; returns the method and a confidence, or the reason for a miss. Never guesses.

輸入結構描述

{
  "type": "object",
  "properties": {
    "name": {
      "type": "string",
      "description": "Organization name as written"
    }
  },
  "required": [
    "name"
  ]
}
🟢get_entity(ticker, entity_id, name)

One company in the federal record: contracts won, PACs, patents, private raises, members of Congress who hold board seats there, insider trades, and its connections graph. Look up by ticker, entity_id, or a name (resolved exactly).

輸入結構描述

{
  "type": "object",
  "properties": {
    "ticker": {
      "type": "string",
      "description": "Stock ticker, e.g. LMT"
    },
    "entity_id": {
      "type": "string",
      "description": "HillSignal entity id (hs:entity:...)"
    },
    "name": {
      "type": "string",
      "description": "Company name; resolved by exact match"
    }
  }
}
🟢search_awards(recipient, ticker, agency, keyword, since, ...)

Search federal contract awards (USAspending, $10M+) by recipient, ticker, awarding agency, keyword, date range or minimum amount. Returns total_matching and pages with offset (up to 50 per call), sorted by date or amount. For "who won the most" questions use award_totals instead, which counts every match.

輸入結構描述

{
  "type": "object",
  "properties": {
    "recipient": {
      "type": "string",
      "description": "Recipient or parent name (partial)"
    },
    "ticker": {
      "type": "string",
      "description": "Resolved ticker"
    },
    "agency": {
      "type": "string",
      "description": "Awarding agency or sub-agency (partial), e.g. \"Navy\""
    },
    "keyword": {
      "type": "string",
      "description": "Word in the award description or NAICS description"
    },
    "since": {
      "type": "string",
      "description": "ISO date (YYYY-MM-DD); only rows on or after this date"
    },
    "until": {
      "type": "string",
      "description": "ISO date; only rows on or before this date"
    },
    "min_amount": {
      "type": "number",
      "description": "Minimum award amount in USD"
    },
    "sort": {
      "type": "string",
      "description": "date (default) | amount"
    },
    "offset": {
      "type": "number",
      "description": "Row offset for paging (default 0)"
    },
    "limit": {
      "type": "number",
      "description": "Max rows, 1-50 (default 25)"
    }
  }
}
🟢award_totals(group_by, recipient, ticker, agency, keyword, ...)

Aggregate every matching federal award (same filters as search_awards) into totals per recipient, per resolved ticker, or per awarding agency: award count, total dollars, share, first and last award, largest award. Use this for "who won the most from the Navy this year" and similar questions; it counts all matches, not a page.

輸入結構描述

{
  "type": "object",
  "properties": {
    "group_by": {
      "type": "string",
      "description": "recipient (default) | ticker | agency"
    },
    "recipient": {
      "type": "string",
      "description": "Recipient or parent name (partial)"
    },
    "ticker": {
      "type": "string",
      "description": "Resolved ticker"
    },
    "agency": {
      "type": "string",
      "description": "Awarding agency or sub-agency (partial)"
    },
    "keyword": {
      "type": "string",
      "description": "Word in the award description"
    },
    "since": {
      "type": "string",
      "description": "ISO date (YYYY-MM-DD); only rows on or after this date"
    },
    "until": {
      "type": "string",
      "description": "ISO date; only rows on or before this date"
    },
    "min_amount": {
      "type": "number",
      "description": "Minimum award amount in USD"
    },
    "limit": {
      "type": "number",
      "description": "Top groups to return, 1-50 (default 25)"
    }
  }
}
🟢search_opportunities(keyword, naics, agency, notice_type, since, ...)

Search SAM.gov notices: Sources Sought and Pre-solicitations (the earliest public signal of a purchase), solicitations and combined synopses. Filter by keyword, NAICS, agency, notice type, date.

輸入結構描述

{
  "type": "object",
  "properties": {
    "keyword": {
      "type": "string",
      "description": "Word in the notice title"
    },
    "naics": {
      "type": "string",
      "description": "NAICS code or prefix, e.g. 3364"
    },
    "agency": {
      "type": "string",
      "description": "Department or organization (partial)"
    },
    "notice_type": {
      "type": "string",
      "description": "sources_sought | presolicitation | solicitation | combined"
    },
    "since": {
      "type": "string",
      "description": "ISO date (YYYY-MM-DD); only rows on or after this date"
    },
    "limit": {
      "type": "number",
      "description": "Max rows, 1-25 (default 10)"
    }
  }
}
🟢congressional_trades(politician, ticker, transaction_type, since, limit)

Stock trades disclosed by members of Congress (House and Senate Periodic Transaction Reports), by member name, ticker, transaction type or date. Returns one row per transaction with the disclosed amount range and the filing PDF. limit caps the FILINGS read, not the trades returned, and filings_matching is how many exist: when truncated is set, far more trades exist than are shown and returned_trades must never be quoted as a total.

輸入結構描述

{
  "type": "object",
  "properties": {
    "politician": {
      "type": "string",
      "description": "Member name (partial)"
    },
    "ticker": {
      "type": "string",
      "description": "Ticker traded"
    },
    "transaction_type": {
      "type": "string",
      "description": "purchase | sale"
    },
    "since": {
      "type": "string",
      "description": "ISO date (YYYY-MM-DD); only rows on or after this date"
    },
    "limit": {
      "type": "number",
      "description": "Max rows, 1-25 (default 10)"
    }
  }
}
🟢get_politician(name, bioguide_id)

A member of Congress: committees, board seats and other positions from annual Financial Disclosures (with resolved tickers), recent stock trades, and top PAC contributors from FEC data. Look up by name or bioguide_id; ambiguous names return candidates.

輸入結構描述

{
  "type": "object",
  "properties": {
    "name": {
      "type": "string",
      "description": "Member name"
    },
    "bioguide_id": {
      "type": "string",
      "description": "Bioguide id, e.g. P000197"
    }
  }
}
🟢insider_trades(ticker, insider, type, min_value, since, ...)

SEC Form 4 insider transactions by ticker or insider name; filter to buys or sells, minimum value, date.

輸入結構描述

{
  "type": "object",
  "properties": {
    "ticker": {
      "type": "string",
      "description": "Ticker"
    },
    "insider": {
      "type": "string",
      "description": "Insider name (partial)"
    },
    "type": {
      "type": "string",
      "description": "buy | sell"
    },
    "min_value": {
      "type": "number",
      "description": "Minimum total value USD"
    },
    "since": {
      "type": "string",
      "description": "ISO date (YYYY-MM-DD); only rows on or after this date"
    },
    "limit": {
      "type": "number",
      "description": "Max rows, 1-25 (default 10)"
    }
  }
}
🟢search_bills(query, bill_number, ticker, sector, event_type, ...)

Full-text search over bills, contract awards and other government events with their analysed sector and ticker exposure. Filter by bill number, ticker, sector, date or event type.

輸入結構描述

{
  "type": "object",
  "properties": {
    "query": {
      "type": "string",
      "description": "Search words"
    },
    "bill_number": {
      "type": "string",
      "description": "e.g. HR1234 or S567"
    },
    "ticker": {
      "type": "string",
      "description": "Ticker named in the analysis"
    },
    "sector": {
      "type": "string",
      "description": "Sector label, e.g. Defense"
    },
    "event_type": {
      "type": "string",
      "description": "Comma-separated event types"
    },
    "since": {
      "type": "string",
      "description": "ISO date (YYYY-MM-DD); only rows on or after this date"
    },
    "limit": {
      "type": "number",
      "description": "Max rows, 1-25 (default 10)"
    }
  }
}
🟢executive_orders(query, sector, since, limit)

Executive orders, proclamations and memoranda with affected sectors and tickers.

輸入結構描述

{
  "type": "object",
  "properties": {
    "query": {
      "type": "string",
      "description": "Words in the title or summary"
    },
    "sector": {
      "type": "string",
      "description": "Sector label"
    },
    "since": {
      "type": "string",
      "description": "ISO date (YYYY-MM-DD); only rows on or after this date"
    },
    "limit": {
      "type": "number",
      "description": "Max rows, 1-25 (default 10)"
    }
  }
}
🟢pac_lookup(ticker, committee_name, limit)

Corporate and trade-association PACs linked to a public company (by ticker) or found by committee name, with the candidates they fund. Answers "does Lockheed have a PAC and who does it give to", "which politicians take money from Pfizer's PAC". For super PACs and mega-donors use super_pacs and money_trail instead.

輸入結構描述

{
  "type": "object",
  "properties": {
    "ticker": {
      "type": "string",
      "description": "Issuer ticker"
    },
    "committee_name": {
      "type": "string",
      "description": "PAC or sponsor name (partial)"
    },
    "limit": {
      "type": "number",
      "description": "Max rows, 1-25 (default 10)"
    }
  }
}
🟢pac_totals(mode, cycle, query, limit)

Which PACs give the most: contributors to tracked candidates ranked by total dollars (FEC Schedule A), by election cycle, with donor-processing conduits like ActBlue and WinRed separated out. mode=disbursers ranks committees by their own reported disbursements instead. Use for "which PACs donate the most to Congress" questions; for super PACs use super_pacs.

輸入結構描述

{
  "type": "object",
  "properties": {
    "mode": {
      "type": "string",
      "description": "contributors (default) | disbursers"
    },
    "cycle": {
      "type": "number",
      "description": "Two-year election cycle, e.g. 2026 or 2024; omit for all"
    },
    "query": {
      "type": "string",
      "description": "Filter contributor or committee name (partial)"
    },
    "limit": {
      "type": "number",
      "description": "Top rows, 1-50 (default 10)"
    }
  }
}
🟢lobbying(ticker, company, bill, since_year, limit)

Senate LDA lobbying disclosures: what a company spent lobbying, which firms it hired, which issues it worked, which agencies it contacted, which bills it named, and how many of its lobbyists previously held government posts. Or pass bill= for everyone who lobbied a given bill. This is the intent layer — super PAC money says who a company is FOR, lobbying says what it WANTS. Answers "what is Coinbase lobbying on", "who lobbied HR1234", "which agencies does Lockheed contact".

輸入結構描述

{
  "type": "object",
  "properties": {
    "ticker": {
      "type": "string",
      "description": "Issuer ticker, e.g. COIN"
    },
    "company": {
      "type": "string",
      "description": "Company name; resolved by exact match"
    },
    "bill": {
      "type": "string",
      "description": "Bill number, e.g. HR1234 or S.567 — returns everyone who lobbied it"
    },
    "since_year": {
      "type": "number",
      "description": "Only filings from this year onward"
    },
    "limit": {
      "type": "number",
      "description": "Max rows, 1-50 (default 20)"
    }
  }
}
🟢money_trail(name, cycle, employer, limit)

How much a person or organization gave, through which entities, to which committees, and what those committees did with it. Answers "how much did Elon Musk give to Trump", "who did Timothy Mellon fund", "where did Uline's money go". Returns donor_identity (the resolved person or organization, every employer spelling folded into one, with lifetime totals), given_via (the LLCs, trusts and partnerships the cheques were actually written under — how the money reached the committee), the recipient committees, and each super PAC's independent spending for and against candidates. Names go in as written ("Elon Musk"); the FEC form ("MUSK, ELON") is handled. Ambiguous names return matched_as; narrow with employer.

輸入結構描述

{
  "type": "object",
  "properties": {
    "name": {
      "type": "string",
      "description": "Donor: a person (\"Ken Griffin\") or organization (\"Uline\")"
    },
    "cycle": {
      "type": "number",
      "description": "Two-year election cycle, e.g. 2024 or 2026; omit for current and previous"
    },
    "employer": {
      "type": "string",
      "description": "Narrow to receipts whose employer field contains this (e.g. \"Tesla\")"
    },
    "limit": {
      "type": "number",
      "description": "Max recipient committees, 1-25 (default 10)"
    }
  },
  "required": [
    "name"
  ]
}
🟢super_pacs(query, committee_id, cycle, sort, types, ...)

Which super PACs are pulling the strings: the largest by money raised, spent or on hand for a cycle, from the FEC's own committee totals. With query or committee_id, one committee's profile: its top donors, what it spent for and against which candidates, and the vendors it paid for campaign work. With shared_vendors, the media buyers and consultancies working for more than one committee. With public_companies, listed companies ranked by the political money they give, resolved to tickers. A committee profile also names the public companies behind its money. Answers "biggest super PACs this cycle", "who funds MAGA Inc", "what is America PAC spending on", "which firms work for both sides", "which public companies give the most".

輸入結構描述

{
  "type": "object",
  "properties": {
    "query": {
      "type": "string",
      "description": "Committee name (partial), e.g. \"America PAC\""
    },
    "committee_id": {
      "type": "string",
      "description": "FEC committee id, e.g. C00879510"
    },
    "cycle": {
      "type": "number",
      "description": "Two-year election cycle (default: current)"
    },
    "sort": {
      "type": "string",
      "description": "receipts (default) | independent_expenditures | disbursements | cash_on_hand"
    },
    "types": {
      "type": "string",
      "description": "\"all\" to include every PAC type; default super PACs and hybrids only"
    },
    "shared_vendors": {
      "type": "boolean",
      "description": "Return vendors paid by two or more committees instead of a committee ranking"
    },
    "min_committees": {
      "type": "number",
      "description": "With shared_vendors: minimum committees sharing a vendor (default 2)"
    },
    "public_companies": {
      "type": "boolean",
      "description": "Return publicly traded companies ranked by political money given, resolved to tickers"
    },
    "ticker": {
      "type": "string",
      "description": "With public_companies: one issuer, e.g. COIN"
    },
    "limit": {
      "type": "number",
      "description": "Max rows, 1-25 (default 10)"
    }
  }
}
🟢outside_money(candidate, cycle, limit)

Outside money on a candidate: every super PAC and other filer spending independently to support or oppose them, with totals, from FEC Schedule E. Answers "who is backing Trump", "which groups are attacking this senator", "how much outside money is in the Ohio Senate race". Candidate names go in as written.

輸入結構描述

{
  "type": "object",
  "properties": {
    "candidate": {
      "type": "string",
      "description": "Candidate name, e.g. \"Donald Trump\", \"Ossoff\""
    },
    "cycle": {
      "type": "number",
      "description": "Two-year election cycle, e.g. 2024 or 2026; omit for all"
    },
    "limit": {
      "type": "number",
      "description": "Max committees per side, 1-50 (default 15)"
    }
  },
  "required": [
    "candidate"
  ]
}
🟢data_coverage

Row counts and the latest dated row for every dataset behind these tools, plus entity-resolution coverage. Call it when an answer depends on how complete the data is, or before saying something does not exist.

輸入結構描述

{
  "type": "object",
  "properties": {}
}
🟢list_themes

The themes the government is currently funding across bills, contracts, solicitations and filings, each with a permanent page and a verdict.

輸入結構描述

{
  "type": "object",
  "properties": {}
}
🟢get_theme(theme_id)

One theme in full: the setup, the dated chronology with each beat linked to its source record, why it matters, what to watch, and the evidence.

輸入結構描述

{
  "type": "object",
  "properties": {
    "theme_id": {
      "type": "string",
      "description": "Theme id from list_themes (with or without hs:theme: prefix)"
    }
  },
  "required": [
    "theme_id"
  ]
}
🟢latest_events(categories, tickers, sources, limit)

The newest events across every source (awards, trades, filings, bills, orders, themes), with provenance. Optional filters by category, ticker or source id.

輸入結構描述

{
  "type": "object",
  "properties": {
    "categories": {
      "type": "string",
      "description": "Comma-separated categories, e.g. contract,trade,filing"
    },
    "tickers": {
      "type": "string",
      "description": "Comma-separated tickers"
    },
    "sources": {
      "type": "string",
      "description": "Comma-separated source ids from /api/wire"
    },
    "limit": {
      "type": "number",
      "description": "Max events, 1-25 (default 20)"
    }
  }
}
🟢track_record

Every forward call HillSignal made since May 2025, verified against SPY 30 days later, with the inclusion rules and the miss rate.

輸入結構描述

{
  "type": "object",
  "properties": {}
}
🟢institutional_holders(ticker, limit)

Largest institutional holders of a ticker from 13F-HR filings HillSignal has ingested, latest position per institution.

輸入結構描述

{
  "type": "object",
  "properties": {
    "ticker": {
      "type": "string",
      "description": "Ticker"
    },
    "limit": {
      "type": "number",
      "description": "Max rows, 1-25 (default 10)"
    }
  },
  "required": [
    "ticker"
  ]
}
🟢foreign_agents(top, rate, country, bill, limit)

Who is paid to represent a foreign government or party inside the United States, and HOW MUCH, from the Justice Department's FARA register and the contracts filed with it. Pass top="contracts" for the biggest deals — the largest contract prices, who pays whom, and who owns the client in the filer's own sworn words. Pass country= to LIST THAT COUNTRY'S AGENTS BY NAME: registered_agents is the full current roster, each with its foreign principal, registration number and dates, plus former_agents and the counts. Answer "how many agents does Israel have and who are they" from that roster. Pass bill= for the foreign interests that lobbied a bill, or nothing for the country leaderboard. A country response also carries also_in_lobbying_data, the minority of those agents that additionally file under the Lobbying Disclosure Act, split into DIRECT, where the lobbying client IS the foreign principal and bills are named, and SHARED_FIRM, where one firm holds a foreign registration and separate domestic clients — context, never a claim about those clients. That subset is never the roster. Countries are spelled as the register spells them. A country roster also carries the MONEY per agent: agreed_price is the contract price from the agreement exhibit filed at registration, always with the rate it is stated at, and reported_receipts is what the agent later reported actually receiving, as a floor rather than earnings. Both are null where no figure could be proved. Each agent may also carry foreign_principal_owned_by — the filer's sworn words on who owns the foreign principal, which is how a US-registered company names its foreign parent.

輸入結構描述

{
  "type": "object",
  "properties": {
    "top": {
      "type": "string",
      "description": "Pass \"contracts\" to rank the BIGGEST foreign-agent contract prices on file — who is paid the most, by whom. Combine with country= to narrow it."
    },
    "rate": {
      "type": "string",
      "description": "With top=contracts: the rate to rank, \"month\" (default) or \"year\", \"quarter\", \"hour\", \"one-time\". Rates are never mixed or converted."
    },
    "country": {
      "type": "string",
      "description": "Country as the register spells it, e.g. \"ISRAEL\", \"SAUDI ARABIA\", \"KOREA, SOUTH\""
    },
    "bill": {
      "type": "string",
      "description": "Bill number, e.g. HR1234 — returns foreign principals that lobbied it"
    },
    "limit": {
      "type": "number",
      "description": "Max rows, 1-100 (default 20)"
    }
  }
}
🟢form_d_raises(query, ticker, since, min_amount, limit)

SEC Form D private placements by issuer name, resolved ticker, date or minimum amount. No minimum size is applied: pass min_amount to impose one. A null offering_amount means the filer stated no offering size, which Form D permits, and is not a raise of zero.

輸入結構描述

{
  "type": "object",
  "properties": {
    "query": {
      "type": "string",
      "description": "Issuer name (partial)"
    },
    "ticker": {
      "type": "string",
      "description": "Resolved public parent ticker"
    },
    "since": {
      "type": "string",
      "description": "ISO date (YYYY-MM-DD); only rows on or after this date"
    },
    "min_amount": {
      "type": "number",
      "description": "Minimum offering amount USD"
    },
    "limit": {
      "type": "number",
      "description": "Max rows, 1-25 (default 10)"
    }
  }
}

社群

為此伺服器評分

證據

近期觀測

已驗證未記錄版本24 個工具
已驗證未記錄版本24 個工具