← 返回 Skills 市场
chuxo

Epstein Files Search

作者 ChuXo · GitHub ↗ · v1.1.0
cross-platform ✓ 安全检测通过
1153
总下载
2
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install epstein
功能描述
Search 44,886+ DOJ-released Jeffrey Epstein documents (Jan 2026 release). Free, no payment required. Search by name, topic, location, or keyword across the full DugganUSA index of declassified Epstein files. Returns document previews, people mentioned, locations, aircraft, evidence types, and source references.
使用说明 (SKILL.md)

Epstein Files Search — Free DOJ Document Search

Search 44,886+ declassified Jeffrey Epstein documents released by the U.S. Department of Justice on January 30, 2026. Powered by the DugganUSA public index.

100% free. No API keys. No accounts. No payment.

Quick Start

# Search by name
node scripts/epstein.mjs search --query "Ghislaine Maxwell" --limit 10

# Search by topic
node scripts/epstein.mjs search --query "flight logs" --limit 20

# Search by location
node scripts/epstein.mjs search --query "Little St James"

# Get index statistics
node scripts/epstein.mjs stats

Commands

search — Search Epstein Documents

Search across all 44,886+ indexed documents by keyword, name, topic, or location.

node scripts/epstein.mjs search --query "SEARCH TERMS" [--limit N]
Flag Description Default
--query \x3Cterms> Search query (required)
--limit \x3CN> Number of results (1-500) 10

Examples:

# Search for a specific person
node scripts/epstein.mjs search --query "Prince Andrew"

# Search for a topic
node scripts/epstein.mjs search --query "financial transactions"

# Search for locations
node scripts/epstein.mjs search --query "New York mansion"

# Get more results
node scripts/epstein.mjs search --query "flight logs" --limit 50

# Search for evidence types
node scripts/epstein.mjs search --query "phone records"

stats — Index Statistics

Get the current state of the document index — total documents, database size, and last update time.

node scripts/epstein.mjs stats

Output Format

Search results are returned as JSON to stdout (for easy piping and parsing). Status messages and Quick Links (direct PDF URLs) go to stderr for easy viewing.

Search Result Shape

{
  "query": "flight logs",
  "totalHits": 1523,
  "hits": [
    {
      "id": "doc-abc123",
      "efta_id": "EFTA-00001234",
      "content_preview": "Excerpt from the document...",
      "doc_type": "legal_document",
      "dataset": "epstein_files",
      "pages": 3,
      "people": ["Person A", "Person B"],
      "locations": ["New York", "Palm Beach"],
      "aircraft": ["N908JE"],
      "evidence_types": ["financial_record"],
      "source": "DOJ Release Jan 2026",
      "indexed_at": "2026-01-31T...",
      "doj_url": "https://www.justice.gov/epstein/files/DataSet%209/EFTA-00001234.pdf",
      "doj_listing_url": "https://www.justice.gov/epstein/doj-disclosures/data-set-9-files"
    }
  ]
}

New in v1.1.0: Each result now includes doj_url (direct PDF link) and doj_listing_url (dataset page). The CLI also displays Quick Links in stderr output:

--- Quick Links ---
1. EFTA-00001234: https://www.justice.gov/epstein/files/DataSet%209/EFTA-00001234.pdf
2. EFTA-00001235: https://www.justice.gov/epstein/files/DataSet%209/EFTA-00001235.pdf

Stats Shape

{
  "totalDocuments": 44886,
  "databaseSize": "2.1 GB",
  "lastUpdate": "2026-01-31T...",
  "isIndexing": false
}

Data Source

All documents come from the U.S. Department of Justice release of Jeffrey Epstein-related records on January 30, 2026. The documents are indexed and searchable via the DugganUSA public API.

  • Source: DOJ Epstein Records
  • Index: DugganUSA Analytics
  • Coverage: 44,886+ document files (3+ million pages)
  • Content: Court filings, depositions, flight logs, financial records, communications, evidence inventories, and more

Piping & Integration

Results go to stdout as JSON, making it easy to pipe into other tools:

# Pipe to jq for filtering
node scripts/epstein.mjs search --query "Maxwell" --limit 100 | jq '.hits[] | .people'

# Save results to file
node scripts/epstein.mjs search --query "flight logs" --limit 500 > flight-logs.json

# Count total hits
node scripts/epstein.mjs search --query "Palm Beach" | jq '.totalHits'

# Extract all mentioned people
node scripts/epstein.mjs search --query "2005" --limit 100 | jq '[.hits[].people[]?] | unique'

Troubleshooting

"Cannot reach API" Check your internet connection. The DugganUSA API may have temporary downtime.

"No results found" Try broader search terms. The search is keyword-based — use names, locations, or document types rather than full sentences.

Slow responses The API typically responds in 100-900ms. Larger result sets (limit > 100) may take slightly longer.

References

安全使用建议
This skill is small and self-contained: it simply calls the DugganUSA analytics API and returns results with DOJ PDF links. Before installing, consider: 1) Confirm you trust the DugganUSA endpoint (https://analytics.dugganusa.com) and its data provenance; the skill will make outbound HTTPS requests to that host. 2) The SKILL.md lists 'curl' but the script uses Node's fetch — ensure you have a recent Node version (Node 18+ recommended for global fetch and AbortSignal.timeout). 3) If you need strong supply-chain assurances, verify the registry/source identity (owner slug is present but homepage/source are sparse). 4) Run the script in a restricted environment if you want to observe network calls first. No credentials or local files are accessed by the code, so there is no obvious secret exfiltration path.
功能分析
Type: OpenClaw Skill Name: epstein Version: 1.1.0 The skill is benign. It clearly states its purpose: searching DOJ-released Jeffrey Epstein documents via the DugganUSA public API. The `scripts/epstein.mjs` code makes HTTP GET requests to `https://analytics.dugganusa.com/api/v1` as documented, processes the JSON response, and outputs results to stdout and stderr. There is no evidence of data exfiltration, malicious execution, persistence mechanisms, prompt injection attempts against the agent in `SKILL.md`, or obfuscation. The `package.json` lists no external dependencies, further reducing supply chain risk.
能力评估
Purpose & Capability
Name/description match the included code: the script queries https://analytics.dugganusa.com/api/v1 for an 'epstein_files' index and returns DOJ PDF links. Minor inconsistencies: the SKILL.md metadata lists a homepage (https://emc2ai.io) while the registry metadata earlier indicated none; SKILL.md declares required binaries 'node' and 'curl' but the provided script only uses node/fetch (curl is not invoked). These are likely harmless but unnecessary.
Instruction Scope
Runtime instructions are limited and explicit (run node scripts/epstein.mjs search|stats). The script only reads CLI args, makes HTTPS requests to the DugganUSA API, and writes JSON to stdout / human lines to stderr. It does not read arbitrary local files, environment variables, or other system configuration.
Install Mechanism
No install spec — instruction-only with a single included script. Nothing is downloaded or written to disk beyond the provided files, which reduces risk.
Credentials
No credentials or environment variables are requested or accessed. The lack of secrets is proportionate to the stated purpose. (Note: the declared 'curl' requirement is unnecessary for the provided Node script.)
Persistence & Privilege
The skill does not request permanent presence (always: false) and does not modify other skills or system configuration. It only runs when invoked.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install epstein
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /epstein 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.1.0
Added direct PDF links (doj_url) to all search results and Quick Links section in CLI output for easy access to DOJ documents
v1.0.0
- Initial release of the Epstein Files Search skill (v1.0.0). - Search 44,886+ DOJ-released Jeffrey Epstein documents by name, topic, location, or keyword. - Returns document previews, people, locations, aircraft, evidence types, and source info. - Includes commands for searching documents and viewing index statistics. - 100% free access, no accounts or payments required. - All results output as JSON for easy scripting and integration.
元数据
Slug epstein
版本 1.1.0
许可证
累计安装 0
当前安装数 0
历史版本数 2
常见问题

Epstein Files Search 是什么?

Search 44,886+ DOJ-released Jeffrey Epstein documents (Jan 2026 release). Free, no payment required. Search by name, topic, location, or keyword across the full DugganUSA index of declassified Epstein files. Returns document previews, people mentioned, locations, aircraft, evidence types, and source references. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 1153 次。

如何安装 Epstein Files Search?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install epstein」即可一键安装,无需额外配置。

Epstein Files Search 是免费的吗?

是的,Epstein Files Search 完全免费(开源免费),可自由下载、安装和使用。

Epstein Files Search 支持哪些平台?

Epstein Files Search 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Epstein Files Search?

由 ChuXo(@chuxo)开发并维护,当前版本 v1.1.0。

💬 留言讨论