# Send Jmail World to your agent Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually. ## Fast path - Download the package from Yavira. - Extract it into a folder your agent can access. - Paste one of the prompts below and point your agent at the extracted folder. ## Suggested prompts ### New install ```text I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete. ``` ### Upgrade existing ```text I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run. ``` ## Machine-readable fields ```json { "schemaVersion": "1.0", "item": { "slug": "jmail", "name": "Jmail World", "source": "tencent", "type": "skill", "category": "开发工具", "sourceUrl": "https://clawhub.ai/fabiantaboo/jmail", "canonicalUrl": "https://clawhub.ai/fabiantaboo/jmail", "targetPlatform": "OpenClaw" }, "install": { "downloadUrl": "/downloads/jmail", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=jmail", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "packageFormat": "ZIP package", "primaryDoc": "SKILL.md", "includedAssets": [ "SKILL.md", "references/api-docs.md", "scripts/jmail-duckdb.sh", "scripts/jmail-search.sh" ], "downloadMode": "redirect", "sourceHealth": { "source": "tencent", "slug": "jmail", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-05-04T13:32:47.391Z", "expiresAt": "2026-05-11T13:32:47.391Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=jmail", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=jmail", "contentDisposition": "attachment; filename=\"jmail-1.6.5.zip\"", "redirectLocation": null, "bodySnippet": null, "slug": "jmail" }, "scope": "item", "summary": "Item download looks usable.", "detail": "Yavira can redirect you to the upstream package for this item.", "primaryActionLabel": "Download for OpenClaw", "primaryActionHref": "/downloads/jmail" }, "validation": { "installChecklist": [ "Use the Yavira download entry.", "Review SKILL.md after the package is downloaded.", "Confirm the extracted package contains the expected setup assets." ], "postInstallChecks": [ "Confirm the extracted package includes the expected docs or setup files.", "Validate the skill or prompts are available in your target agent workspace.", "Capture any manual follow-up steps the agent could not complete." ] } }, "links": { "detailUrl": "https://openagent3.xyz/skills/jmail", "downloadUrl": "https://openagent3.xyz/downloads/jmail", "agentUrl": "https://openagent3.xyz/skills/jmail/agent", "manifestUrl": "https://openagent3.xyz/skills/jmail/agent.json", "briefUrl": "https://openagent3.xyz/skills/jmail/agent.md" } } ``` ## Documentation ### jmail-world — Epstein Archive Search & Analysis Search and analyze Jeffrey Epstein's email archive, iMessages, documents, photos, and people on jmail.world. ### When to Use Researching Epstein connections and communications Finding emails between specific people Exploring iMessage conversations Analyzing communication networks and timelines Looking up people identified in photos Searching documents (DOJ releases, court records) ### 1. Web Search API (quick lookups) GET https://jmail.world/api/emails/search?q=QUERY&limit=50&page=1&source=all&from=NAME No auth needed. Use web_fetch or curl. Good for quick keyword searches. ### 2. DuckDB + Parquet (deep analysis) All datasets served as static Parquet files from https://data.jmail.world/v1/. No API key, no rate limit, no auth. Use the included scripts: # Full-text email search (Web API) bash scripts/jmail-search.sh "query text" bash scripts/jmail-search.sh "scopolamine" --from "Epstein" --limit 20 # DuckDB queries (local Parquet, no rate limits) bash scripts/jmail-duckdb.sh search "flight manifest" bash scripts/jmail-duckdb.sh person "Ghislaine Maxwell" bash scripts/jmail-duckdb.sh imessages "ghislaine-maxwell" bash scripts/jmail-duckdb.sh imessage-search "AfD" bash scripts/jmail-duckdb.sh imessage-search "Merkel" --from "Bannon" bash scripts/jmail-duckdb.sh imessage-list bash scripts/jmail-duckdb.sh network "Bill Clinton" bash scripts/jmail-duckdb.sh timeline "2005-01-01" "2005-12-31" bash scripts/jmail-duckdb.sh top-senders bash scripts/jmail-duckdb.sh people bash scripts/jmail-duckdb.sh documents "flight" bash scripts/jmail-duckdb.sh photos "person-name" bash scripts/jmail-duckdb.sh photo-search "pool" bash scripts/jmail-duckdb.sh photo-download "EFTA00000002-0.png" ./output bash scripts/jmail-duckdb.sh download "EFTA02406146" bash scripts/jmail-duckdb.sh download "HOUSE_OVERSIGHT_034601" bash scripts/jmail-duckdb.sh stars ### Available Datasets DatasetURLSizeRecordsEmails (full)emails.parquet334MB1.78MEmails (slim)emails-slim.parquet41MB1.78MDocumentsdocuments.parquet25MB1.41MPhotosphotos.parquet~1MB18KPeoplepeople.parquet<100KB473Photo Facesphoto_faces.parquet<100KB975iMessage Conversationsimessage_conversations.parquet——iMessage Messagesimessage_messages.parquet——Star Countsstar_counts.parquet~2MB414KRelease Batchesrelease_batches.parquet<10KB— All under https://data.jmail.world/v1/. ### Email Columns (slim) id, doc_id, sender, subject, to_recipients (json), cc_recipients (json), bcc_recipients (json), sent_at (timestamp), account_email, email_drop_id, epstein_is_sender (bool) ### Email Additional (full) content_markdown, content_html, attachments (int) ### iMessage Conversations id, slug, name, bio, photo, last_message, last_message_time, pinned, confirmed, source_files (json), message_count ### iMessage Messages id, conversation_slug, message_index, text, sender ("me" = Epstein, "them" = contact), time, timestamp, source_file, sender_name ### People id, name, source, photo_count ### Documents id, source, release_batch, original_filename, page_count, size, document_description, has_thumbnail ### Web Pages (browser needed) /person/SLUG — Person profile /flights — Flight records /photos — Photo browser /drive/new-only — New documents /topic/SLUG — Topic pages ### Document Full-Text Search Documents have sharded full-text files (large downloads): documents-full/VOL00008.parquet — DOJ Volume 8 documents-full/VOL00009.parquet — DOJ Volume 9 documents-full/VOL00010.parquet — DOJ Volume 10 documents-full/DataSet11.parquet — DOJ Dataset 11 documents-full/other.parquet — House Oversight, court records Query with DuckDB: SELECT id, original_filename, extracted_text FROM read_parquet('https://data.jmail.world/v1/documents-full/other.parquet') WHERE extracted_text ILIKE '%rothschild%' LIMIT 10; ### Photo Columns (full schema) id, source, release_batch, original_filename, content_type (MIME), width (px), height (px), image_description (AI-generated) Search photos by description: SELECT original_filename, image_description, width, height FROM read_parquet('https://data.jmail.world/v1/photos.parquet') WHERE image_description ILIKE '%pool%' LIMIT 20; ### Universal Download (any document, email, or photo) # Download by document ID — auto-detects type and source bash scripts/jmail-duckdb.sh download "EFTA02406146" # DOJ email PDF bash scripts/jmail-duckdb.sh download "HOUSE_OVERSIGHT_034601" # House Oversight photo bash scripts/jmail-duckdb.sh download "COURT_giuffre-115cv07433_1" # Court document bash scripts/jmail-duckdb.sh download "vol00009-efta00462570-pdf" # DOJ volume scan bash scripts/jmail-duckdb.sh download "EFTA02406146" ./output-dir # Custom output dir Supported sources: ID PatternTypeFormatEFTA*DOJ emails & photosPDFvol*DOJ volume scansPDFHOUSE_OVERSIGHT_*House Oversight photosJPG (direct)COURT_giuffre*Giuffre court docsPDF ### Photo Search & Download # Search photos by AI-generated description bash scripts/jmail-duckdb.sh photo-search "swimming pool" bash scripts/jmail-duckdb.sh photo-search "forced entry" # Download a specific photo (DOJ: extracts PNG from PDF, HO: direct JPG) bash scripts/jmail-duckdb.sh photo-download "EFTA00000002-0.png" bash scripts/jmail-duckdb.sh photo-download "HOUSE_OVERSIGHT_034601.JPG" ./output DOJ photos are stored as single-page PDFs — the download command automatically extracts the embedded image as PNG using pdfimages. House Oversight photos are direct JPGs. If pdfimages is not installed, DOJ photos are saved as PDF. ### Security SQL injection prevention: All user input is whitelist-sanitized (alphanumeric + safe chars only). SQL meta-characters, operators, and keywords are stripped before query construction. Path traversal prevention: Parquet filenames and document IDs are validated against strict patterns before use in file operations. Read-only queries: All DuckDB operations are SELECT on read_parquet() — no writes, no code execution, no network access beyond the cached Parquet files. Data source: All data comes from jmail.world's public Parquet files and assets.getkino.com (DOJ document mirror). No private APIs or credentials involved. ### Requirements DuckDB (duckdb CLI) — must be installed manually (apt install duckdb / brew install duckdb / duckdb.org) curl — for web search API and downloading Parquet files python3 — for URL encoding in search script pdfimages (optional, for photo-download PNG extraction) — apt install poppler-utils jq (optional, for JSON formatting) ### References Official Docs Docs Index for LLMs API Reference DuckDB Examples Python Client (external) — not bundled, see official docs Datasets & URLs ## Trust - Source: tencent - Verification: Indexed source record - Publisher: fabiantaboo - Version: 1.6.5 ## Source health - Status: healthy - Item download looks usable. - Yavira can redirect you to the upstream package for this item. - Health scope: item - Reason: direct_download_ok - Checked at: 2026-05-04T13:32:47.391Z - Expires at: 2026-05-11T13:32:47.391Z - Recommended action: Download for OpenClaw ## Links - [Detail page](https://openagent3.xyz/skills/jmail) - [Send to Agent page](https://openagent3.xyz/skills/jmail/agent) - [JSON manifest](https://openagent3.xyz/skills/jmail/agent.json) - [Markdown brief](https://openagent3.xyz/skills/jmail/agent.md) - [Download page](https://openagent3.xyz/downloads/jmail)